Upload verified mixed BFP4/BFP8 checkpoint with MTP and evaluation evidence
Browse filesThis view is limited to 50 files because it contains too many changes. See raw diff
- .dockerignore +6 -0
- .gitattributes +250 -0
- LICENSE +202 -0
- SHA256SUMS +695 -0
- api_validate.py +125 -0
- build_native_checkpoint.py +277 -0
- calibration/importance.npz +3 -0
- calibration/instruction-calibration.jsonl +0 -0
- calibration/metadata.json +1337 -0
- checkpoint/LICENSE +202 -0
- checkpoint/chat_template.jinja +154 -0
- checkpoint/config.json +103 -0
- checkpoint/equivalence-mtp.json +342 -0
- checkpoint/equivalence.json +342 -0
- checkpoint/merges.txt +0 -0
- checkpoint/native_checkpoint.py +315 -0
- checkpoint/native_manifest.json +0 -0
- checkpoint/preprocessor_config.json +21 -0
- checkpoint/provenance/build_native_checkpoint.py +277 -0
- checkpoint/provenance/importance-metadata.json +1337 -0
- checkpoint/provenance/importance.npz +3 -0
- checkpoint/provenance/original.safetensors.index.json +782 -0
- checkpoint/provenance/overrides.json +73 -0
- checkpoint/provenance/precision-plan.json +233 -0
- checkpoint/provenance/tt_eval.py +440 -0
- checkpoint/provenance/weight_mapping.py +220 -0
- checkpoint/quantization-error.json +0 -0
- checkpoint/tensors/00000.tensorbin +3 -0
- checkpoint/tensors/00001.safetensors +3 -0
- checkpoint/tensors/00002.tensorbin +3 -0
- checkpoint/tensors/00003.tensorbin +3 -0
- checkpoint/tensors/00004.tensorbin +3 -0
- checkpoint/tensors/00005.tensorbin +3 -0
- checkpoint/tensors/00006.tensorbin +3 -0
- checkpoint/tensors/00007.tensorbin +3 -0
- checkpoint/tensors/00008.tensorbin +3 -0
- checkpoint/tensors/00009.tensorbin +3 -0
- checkpoint/tensors/00010.tensorbin +3 -0
- checkpoint/tensors/00011.tensorbin +3 -0
- checkpoint/tensors/00012.tensorbin +3 -0
- checkpoint/tensors/00013.tensorbin +3 -0
- checkpoint/tensors/00014.tensorbin +3 -0
- checkpoint/tensors/00015.tensorbin +3 -0
- checkpoint/tensors/00016.tensorbin +3 -0
- checkpoint/tensors/00017.tensorbin +3 -0
- checkpoint/tensors/00018.tensorbin +3 -0
- checkpoint/tensors/00019.tensorbin +3 -0
- checkpoint/tensors/00020.tensorbin +3 -0
- checkpoint/tensors/00021.tensorbin +3 -0
- checkpoint/tensors/00022.tensorbin +3 -0
.dockerignore
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
*
|
| 2 |
+
!runtime/
|
| 3 |
+
!runtime/**
|
| 4 |
+
!native_checkpoint.py
|
| 5 |
+
**/__pycache__/
|
| 6 |
+
**/*.pyc
|
.gitattributes
CHANGED
|
@@ -33,3 +33,253 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
checkpoint/tensors/00000.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
checkpoint/tensors/00002.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
checkpoint/tensors/00003.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
checkpoint/tensors/00004.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 40 |
+
checkpoint/tensors/00005.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 41 |
+
checkpoint/tensors/00006.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 42 |
+
checkpoint/tensors/00007.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 43 |
+
checkpoint/tensors/00008.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 44 |
+
checkpoint/tensors/00009.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 45 |
+
checkpoint/tensors/00010.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 46 |
+
checkpoint/tensors/00011.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 47 |
+
checkpoint/tensors/00012.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 48 |
+
checkpoint/tensors/00013.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 49 |
+
checkpoint/tensors/00014.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 50 |
+
checkpoint/tensors/00015.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 51 |
+
checkpoint/tensors/00016.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 52 |
+
checkpoint/tensors/00017.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 53 |
+
checkpoint/tensors/00018.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 54 |
+
checkpoint/tensors/00019.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 55 |
+
checkpoint/tensors/00020.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 56 |
+
checkpoint/tensors/00021.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 57 |
+
checkpoint/tensors/00022.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 58 |
+
checkpoint/tensors/00023.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 59 |
+
checkpoint/tensors/00024.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 60 |
+
checkpoint/tensors/00025.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 61 |
+
checkpoint/tensors/00026.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 62 |
+
checkpoint/tensors/00027.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 63 |
+
checkpoint/tensors/00028.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 64 |
+
checkpoint/tensors/00029.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 65 |
+
checkpoint/tensors/00030.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 66 |
+
checkpoint/tensors/00031.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 67 |
+
checkpoint/tensors/00032.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 68 |
+
checkpoint/tensors/00033.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 69 |
+
checkpoint/tensors/00034.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 70 |
+
checkpoint/tensors/00035.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 71 |
+
checkpoint/tensors/00036.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 72 |
+
checkpoint/tensors/00037.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 73 |
+
checkpoint/tensors/00038.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 74 |
+
checkpoint/tensors/00039.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 75 |
+
checkpoint/tensors/00040.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 76 |
+
checkpoint/tensors/00041.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 77 |
+
checkpoint/tensors/00042.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 78 |
+
checkpoint/tensors/00043.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 79 |
+
checkpoint/tensors/00044.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 80 |
+
checkpoint/tensors/00045.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 81 |
+
checkpoint/tensors/00046.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 82 |
+
checkpoint/tensors/00047.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 83 |
+
checkpoint/tensors/00048.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 84 |
+
checkpoint/tensors/00049.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 85 |
+
checkpoint/tensors/00050.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 86 |
+
checkpoint/tensors/00051.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 87 |
+
checkpoint/tensors/00052.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 88 |
+
checkpoint/tensors/00053.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 89 |
+
checkpoint/tensors/00054.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 90 |
+
checkpoint/tensors/00055.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 91 |
+
checkpoint/tensors/00056.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 92 |
+
checkpoint/tensors/00057.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 93 |
+
checkpoint/tensors/00058.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 94 |
+
checkpoint/tensors/00059.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 95 |
+
checkpoint/tensors/00060.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 96 |
+
checkpoint/tensors/00061.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 97 |
+
checkpoint/tensors/00062.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 98 |
+
checkpoint/tensors/00063.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 99 |
+
checkpoint/tensors/00067.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 100 |
+
checkpoint/tensors/00068.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 101 |
+
checkpoint/tensors/00069.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 102 |
+
checkpoint/tensors/00070.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 103 |
+
checkpoint/tensors/00071.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 104 |
+
checkpoint/tensors/00072.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 105 |
+
checkpoint/tensors/00073.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 106 |
+
checkpoint/tensors/00074.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 107 |
+
checkpoint/tensors/00075.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 108 |
+
checkpoint/tensors/00076.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 109 |
+
checkpoint/tensors/00077.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 110 |
+
checkpoint/tensors/00078.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 111 |
+
checkpoint/tensors/00079.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 112 |
+
checkpoint/tensors/00080.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 113 |
+
checkpoint/tensors/00081.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 114 |
+
checkpoint/tensors/00082.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 115 |
+
checkpoint/tensors/00083.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 116 |
+
checkpoint/tensors/00084.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 117 |
+
checkpoint/tensors/00085.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 118 |
+
checkpoint/tensors/00086.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 119 |
+
checkpoint/tensors/00087.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 120 |
+
checkpoint/tensors/00088.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 121 |
+
checkpoint/tensors/00089.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 122 |
+
checkpoint/tensors/00090.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 123 |
+
checkpoint/tensors/00091.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 124 |
+
checkpoint/tensors/00092.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 125 |
+
checkpoint/tensors/00093.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 126 |
+
checkpoint/tensors/00094.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 127 |
+
checkpoint/tensors/00095.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 128 |
+
checkpoint/tensors/00096.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 129 |
+
checkpoint/tensors/00097.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 130 |
+
checkpoint/tensors/00098.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 131 |
+
checkpoint/tensors/00099.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 132 |
+
checkpoint/tensors/00100.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 133 |
+
checkpoint/tensors/00101.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 134 |
+
checkpoint/tensors/00102.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 135 |
+
checkpoint/tensors/00103.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 136 |
+
checkpoint/tensors/00104.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 137 |
+
checkpoint/tensors/00105.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 138 |
+
checkpoint/tensors/00106.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 139 |
+
checkpoint/tensors/00107.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 140 |
+
checkpoint/tensors/00108.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 141 |
+
checkpoint/tensors/00109.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 142 |
+
checkpoint/tensors/00110.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 143 |
+
checkpoint/tensors/00111.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 144 |
+
checkpoint/tensors/00112.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 145 |
+
checkpoint/tensors/00113.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 146 |
+
checkpoint/tensors/00114.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 147 |
+
checkpoint/tensors/00115.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 148 |
+
checkpoint/tensors/00116.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 149 |
+
checkpoint/tensors/00117.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 150 |
+
checkpoint/tensors/00118.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 151 |
+
checkpoint/tensors/00119.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 152 |
+
checkpoint/tensors/00120.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 153 |
+
checkpoint/tensors/00121.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 154 |
+
checkpoint/tensors/00122.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 155 |
+
checkpoint/tensors/00123.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 156 |
+
checkpoint/tensors/00124.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 157 |
+
checkpoint/tensors/00125.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 158 |
+
checkpoint/tensors/00126.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 159 |
+
checkpoint/tensors/00127.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 160 |
+
checkpoint/tensors/00136.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 161 |
+
checkpoint/tensors/00137.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 162 |
+
checkpoint/tensors/00138.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 163 |
+
checkpoint/tensors/00140.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 164 |
+
checkpoint/tensors/00148.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 165 |
+
checkpoint/tensors/00149.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 166 |
+
checkpoint/tensors/00150.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 167 |
+
checkpoint/tensors/00152.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 168 |
+
checkpoint/tensors/00160.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 169 |
+
checkpoint/tensors/00161.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 170 |
+
checkpoint/tensors/00162.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 171 |
+
checkpoint/tensors/00164.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 172 |
+
checkpoint/tensors/00169.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 173 |
+
checkpoint/tensors/00170.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 174 |
+
checkpoint/tensors/00172.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 175 |
+
checkpoint/tensors/00179.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 176 |
+
checkpoint/tensors/00180.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 177 |
+
checkpoint/tensors/00181.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 178 |
+
checkpoint/tensors/00182.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 179 |
+
checkpoint/tensors/00184.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 180 |
+
checkpoint/tensors/00192.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 181 |
+
checkpoint/tensors/00193.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 182 |
+
checkpoint/tensors/00194.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 183 |
+
checkpoint/tensors/00195.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 184 |
+
checkpoint/tensors/00197.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 185 |
+
checkpoint/tensors/00205.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 186 |
+
checkpoint/tensors/00206.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 187 |
+
checkpoint/tensors/00207.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 188 |
+
checkpoint/tensors/00209.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 189 |
+
checkpoint/tensors/00214.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 190 |
+
checkpoint/tensors/00215.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 191 |
+
checkpoint/tensors/00217.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 192 |
+
checkpoint/tensors/00224.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 193 |
+
checkpoint/tensors/00225.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 194 |
+
checkpoint/tensors/00226.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 195 |
+
checkpoint/tensors/00228.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 196 |
+
checkpoint/tensors/00236.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 197 |
+
checkpoint/tensors/00237.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 198 |
+
checkpoint/tensors/00238.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 199 |
+
checkpoint/tensors/00240.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 200 |
+
checkpoint/tensors/00248.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 201 |
+
checkpoint/tensors/00249.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 202 |
+
checkpoint/tensors/00250.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 203 |
+
checkpoint/tensors/00252.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 204 |
+
checkpoint/tensors/00257.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 205 |
+
checkpoint/tensors/00258.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 206 |
+
checkpoint/tensors/00260.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 207 |
+
checkpoint/tensors/00267.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 208 |
+
checkpoint/tensors/00268.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 209 |
+
checkpoint/tensors/00269.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 210 |
+
checkpoint/tensors/00271.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 211 |
+
checkpoint/tensors/00279.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 212 |
+
checkpoint/tensors/00280.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 213 |
+
checkpoint/tensors/00281.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 214 |
+
checkpoint/tensors/00283.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 215 |
+
checkpoint/tensors/00291.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 216 |
+
checkpoint/tensors/00292.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 217 |
+
checkpoint/tensors/00293.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 218 |
+
checkpoint/tensors/00294.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 219 |
+
checkpoint/tensors/00296.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 220 |
+
checkpoint/tensors/00304.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 221 |
+
checkpoint/tensors/00305.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 222 |
+
checkpoint/tensors/00306.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 223 |
+
checkpoint/tensors/00308.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 224 |
+
checkpoint/tensors/00313.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 225 |
+
checkpoint/tensors/00314.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 226 |
+
checkpoint/tensors/00316.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 227 |
+
checkpoint/tensors/00323.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 228 |
+
checkpoint/tensors/00324.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 229 |
+
checkpoint/tensors/00325.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 230 |
+
checkpoint/tensors/00327.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 231 |
+
checkpoint/tensors/00335.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 232 |
+
checkpoint/tensors/00336.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 233 |
+
checkpoint/tensors/00337.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 234 |
+
checkpoint/tensors/00339.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 235 |
+
checkpoint/tensors/00347.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 236 |
+
checkpoint/tensors/00348.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 237 |
+
checkpoint/tensors/00349.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 238 |
+
checkpoint/tensors/00351.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 239 |
+
checkpoint/tensors/00356.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 240 |
+
checkpoint/tensors/00357.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 241 |
+
checkpoint/tensors/00359.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 242 |
+
checkpoint/tensors/00366.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 243 |
+
checkpoint/tensors/00367.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 244 |
+
checkpoint/tensors/00368.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 245 |
+
checkpoint/tensors/00369.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 246 |
+
checkpoint/tensors/00371.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 247 |
+
checkpoint/tensors/00379.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 248 |
+
checkpoint/tensors/00380.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 249 |
+
checkpoint/tensors/00381.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 250 |
+
checkpoint/tensors/00382.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 251 |
+
checkpoint/tensors/00384.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 252 |
+
checkpoint/tensors/00389.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 253 |
+
checkpoint/tensors/00390.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 254 |
+
checkpoint/tensors/00392.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 255 |
+
checkpoint/tensors/00399.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 256 |
+
checkpoint/tensors/00400.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 257 |
+
checkpoint/tensors/00401.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 258 |
+
checkpoint/tensors/00403.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 259 |
+
checkpoint/tensors/00408.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 260 |
+
checkpoint/tensors/00409.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 261 |
+
checkpoint/tensors/00411.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 262 |
+
checkpoint/tensors/00418.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 263 |
+
checkpoint/tensors/00419.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 264 |
+
checkpoint/tensors/00420.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 265 |
+
checkpoint/tensors/00422.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 266 |
+
checkpoint/tensors/00430.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 267 |
+
checkpoint/tensors/00431.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 268 |
+
checkpoint/tensors/00432.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 269 |
+
checkpoint/tensors/00434.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 270 |
+
checkpoint/tensors/00442.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 271 |
+
checkpoint/tensors/00443.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 272 |
+
checkpoint/tensors/00444.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 273 |
+
checkpoint/tensors/00446.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 274 |
+
checkpoint/tensors/00451.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 275 |
+
checkpoint/tensors/00452.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 276 |
+
checkpoint/tensors/00454.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 277 |
+
checkpoint/tensors/00461.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 278 |
+
checkpoint/tensors/00462.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 279 |
+
checkpoint/tensors/00463.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 280 |
+
checkpoint/tensors/00465.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 281 |
+
checkpoint/tensors/00473.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 282 |
+
checkpoint/tensors/00474.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 283 |
+
checkpoint/tensors/00475.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 284 |
+
checkpoint/tensors/00477.tensorbin filter=lfs diff=lfs merge=lfs -text
|
| 285 |
+
checkpoint/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
LICENSE
ADDED
|
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
Apache License
|
| 3 |
+
Version 2.0, January 2004
|
| 4 |
+
http://www.apache.org/licenses/
|
| 5 |
+
|
| 6 |
+
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
|
| 7 |
+
|
| 8 |
+
1. Definitions.
|
| 9 |
+
|
| 10 |
+
"License" shall mean the terms and conditions for use, reproduction,
|
| 11 |
+
and distribution as defined by Sections 1 through 9 of this document.
|
| 12 |
+
|
| 13 |
+
"Licensor" shall mean the copyright owner or entity authorized by
|
| 14 |
+
the copyright owner that is granting the License.
|
| 15 |
+
|
| 16 |
+
"Legal Entity" shall mean the union of the acting entity and all
|
| 17 |
+
other entities that control, are controlled by, or are under common
|
| 18 |
+
control with that entity. For the purposes of this definition,
|
| 19 |
+
"control" means (i) the power, direct or indirect, to cause the
|
| 20 |
+
direction or management of such entity, whether by contract or
|
| 21 |
+
otherwise, or (ii) ownership of fifty percent (50%) or more of the
|
| 22 |
+
outstanding shares, or (iii) beneficial ownership of such entity.
|
| 23 |
+
|
| 24 |
+
"You" (or "Your") shall mean an individual or Legal Entity
|
| 25 |
+
exercising permissions granted by this License.
|
| 26 |
+
|
| 27 |
+
"Source" form shall mean the preferred form for making modifications,
|
| 28 |
+
including but not limited to software source code, documentation
|
| 29 |
+
source, and configuration files.
|
| 30 |
+
|
| 31 |
+
"Object" form shall mean any form resulting from mechanical
|
| 32 |
+
transformation or translation of a Source form, including but
|
| 33 |
+
not limited to compiled object code, generated documentation,
|
| 34 |
+
and conversions to other media types.
|
| 35 |
+
|
| 36 |
+
"Work" shall mean the work of authorship, whether in Source or
|
| 37 |
+
Object form, made available under the License, as indicated by a
|
| 38 |
+
copyright notice that is included in or attached to the work
|
| 39 |
+
(an example is provided in the Appendix below).
|
| 40 |
+
|
| 41 |
+
"Derivative Works" shall mean any work, whether in Source or Object
|
| 42 |
+
form, that is based on (or derived from) the Work and for which the
|
| 43 |
+
editorial revisions, annotations, elaborations, or other modifications
|
| 44 |
+
represent, as a whole, an original work of authorship. For the purposes
|
| 45 |
+
of this License, Derivative Works shall not include works that remain
|
| 46 |
+
separable from, or merely link (or bind by name) to the interfaces of,
|
| 47 |
+
the Work and Derivative Works thereof.
|
| 48 |
+
|
| 49 |
+
"Contribution" shall mean any work of authorship, including
|
| 50 |
+
the original version of the Work and any modifications or additions
|
| 51 |
+
to that Work or Derivative Works thereof, that is intentionally
|
| 52 |
+
submitted to Licensor for inclusion in the Work by the copyright owner
|
| 53 |
+
or by an individual or Legal Entity authorized to submit on behalf of
|
| 54 |
+
the copyright owner. For the purposes of this definition, "submitted"
|
| 55 |
+
means any form of electronic, verbal, or written communication sent
|
| 56 |
+
to the Licensor or its representatives, including but not limited to
|
| 57 |
+
communication on electronic mailing lists, source code control systems,
|
| 58 |
+
and issue tracking systems that are managed by, or on behalf of, the
|
| 59 |
+
Licensor for the purpose of discussing and improving the Work, but
|
| 60 |
+
excluding communication that is conspicuously marked or otherwise
|
| 61 |
+
designated in writing by the copyright owner as "Not a Contribution."
|
| 62 |
+
|
| 63 |
+
"Contributor" shall mean Licensor and any individual or Legal Entity
|
| 64 |
+
on behalf of whom a Contribution has been received by Licensor and
|
| 65 |
+
subsequently incorporated within the Work.
|
| 66 |
+
|
| 67 |
+
2. Grant of Copyright License. Subject to the terms and conditions of
|
| 68 |
+
this License, each Contributor hereby grants to You a perpetual,
|
| 69 |
+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
| 70 |
+
copyright license to reproduce, prepare Derivative Works of,
|
| 71 |
+
publicly display, publicly perform, sublicense, and distribute the
|
| 72 |
+
Work and such Derivative Works in Source or Object form.
|
| 73 |
+
|
| 74 |
+
3. Grant of Patent License. Subject to the terms and conditions of
|
| 75 |
+
this License, each Contributor hereby grants to You a perpetual,
|
| 76 |
+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
| 77 |
+
(except as stated in this section) patent license to make, have made,
|
| 78 |
+
use, offer to sell, sell, import, and otherwise transfer the Work,
|
| 79 |
+
where such license applies only to those patent claims licensable
|
| 80 |
+
by such Contributor that are necessarily infringed by their
|
| 81 |
+
Contribution(s) alone or by combination of their Contribution(s)
|
| 82 |
+
with the Work to which such Contribution(s) was submitted. If You
|
| 83 |
+
institute patent litigation against any entity (including a
|
| 84 |
+
cross-claim or counterclaim in a lawsuit) alleging that the Work
|
| 85 |
+
or a Contribution incorporated within the Work constitutes direct
|
| 86 |
+
or contributory patent infringement, then any patent licenses
|
| 87 |
+
granted to You under this License for that Work shall terminate
|
| 88 |
+
as of the date such litigation is filed.
|
| 89 |
+
|
| 90 |
+
4. Redistribution. You may reproduce and distribute copies of the
|
| 91 |
+
Work or Derivative Works thereof in any medium, with or without
|
| 92 |
+
modifications, and in Source or Object form, provided that You
|
| 93 |
+
meet the following conditions:
|
| 94 |
+
|
| 95 |
+
(a) You must give any other recipients of the Work or
|
| 96 |
+
Derivative Works a copy of this License; and
|
| 97 |
+
|
| 98 |
+
(b) You must cause any modified files to carry prominent notices
|
| 99 |
+
stating that You changed the files; and
|
| 100 |
+
|
| 101 |
+
(c) You must retain, in the Source form of any Derivative Works
|
| 102 |
+
that You distribute, all copyright, patent, trademark, and
|
| 103 |
+
attribution notices from the Source form of the Work,
|
| 104 |
+
excluding those notices that do not pertain to any part of
|
| 105 |
+
the Derivative Works; and
|
| 106 |
+
|
| 107 |
+
(d) If the Work includes a "NOTICE" text file as part of its
|
| 108 |
+
distribution, then any Derivative Works that You distribute must
|
| 109 |
+
include a readable copy of the attribution notices contained
|
| 110 |
+
within such NOTICE file, excluding those notices that do not
|
| 111 |
+
pertain to any part of the Derivative Works, in at least one
|
| 112 |
+
of the following places: within a NOTICE text file distributed
|
| 113 |
+
as part of the Derivative Works; within the Source form or
|
| 114 |
+
documentation, if provided along with the Derivative Works; or,
|
| 115 |
+
within a display generated by the Derivative Works, if and
|
| 116 |
+
wherever such third-party notices normally appear. The contents
|
| 117 |
+
of the NOTICE file are for informational purposes only and
|
| 118 |
+
do not modify the License. You may add Your own attribution
|
| 119 |
+
notices within Derivative Works that You distribute, alongside
|
| 120 |
+
or as an addendum to the NOTICE text from the Work, provided
|
| 121 |
+
that such additional attribution notices cannot be construed
|
| 122 |
+
as modifying the License.
|
| 123 |
+
|
| 124 |
+
You may add Your own copyright statement to Your modifications and
|
| 125 |
+
may provide additional or different license terms and conditions
|
| 126 |
+
for use, reproduction, or distribution of Your modifications, or
|
| 127 |
+
for any such Derivative Works as a whole, provided Your use,
|
| 128 |
+
reproduction, and distribution of the Work otherwise complies with
|
| 129 |
+
the conditions stated in this License.
|
| 130 |
+
|
| 131 |
+
5. Submission of Contributions. Unless You explicitly state otherwise,
|
| 132 |
+
any Contribution intentionally submitted for inclusion in the Work
|
| 133 |
+
by You to the Licensor shall be under the terms and conditions of
|
| 134 |
+
this License, without any additional terms or conditions.
|
| 135 |
+
Notwithstanding the above, nothing herein shall supersede or modify
|
| 136 |
+
the terms of any separate license agreement you may have executed
|
| 137 |
+
with Licensor regarding such Contributions.
|
| 138 |
+
|
| 139 |
+
6. Trademarks. This License does not grant permission to use the trade
|
| 140 |
+
names, trademarks, service marks, or product names of the Licensor,
|
| 141 |
+
except as required for reasonable and customary use in describing the
|
| 142 |
+
origin of the Work and reproducing the content of the NOTICE file.
|
| 143 |
+
|
| 144 |
+
7. Disclaimer of Warranty. Unless required by applicable law or
|
| 145 |
+
agreed to in writing, Licensor provides the Work (and each
|
| 146 |
+
Contributor provides its Contributions) on an "AS IS" BASIS,
|
| 147 |
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
|
| 148 |
+
implied, including, without limitation, any warranties or conditions
|
| 149 |
+
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
|
| 150 |
+
PARTICULAR PURPOSE. You are solely responsible for determining the
|
| 151 |
+
appropriateness of using or redistributing the Work and assume any
|
| 152 |
+
risks associated with Your exercise of permissions under this License.
|
| 153 |
+
|
| 154 |
+
8. Limitation of Liability. In no event and under no legal theory,
|
| 155 |
+
whether in tort (including negligence), contract, or otherwise,
|
| 156 |
+
unless required by applicable law (such as deliberate and grossly
|
| 157 |
+
negligent acts) or agreed to in writing, shall any Contributor be
|
| 158 |
+
liable to You for damages, including any direct, indirect, special,
|
| 159 |
+
incidental, or consequential damages of any character arising as a
|
| 160 |
+
result of this License or out of the use or inability to use the
|
| 161 |
+
Work (including but not limited to damages for loss of goodwill,
|
| 162 |
+
work stoppage, computer failure or malfunction, or any and all
|
| 163 |
+
other commercial damages or losses), even if such Contributor
|
| 164 |
+
has been advised of the possibility of such damages.
|
| 165 |
+
|
| 166 |
+
9. Accepting Warranty or Additional Liability. While redistributing
|
| 167 |
+
the Work or Derivative Works thereof, You may choose to offer,
|
| 168 |
+
and charge a fee for, acceptance of support, warranty, indemnity,
|
| 169 |
+
or other liability obligations and/or rights consistent with this
|
| 170 |
+
License. However, in accepting such obligations, You may act only
|
| 171 |
+
on Your own behalf and on Your sole responsibility, not on behalf
|
| 172 |
+
of any other Contributor, and only if You agree to indemnify,
|
| 173 |
+
defend, and hold each Contributor harmless for any liability
|
| 174 |
+
incurred by, or claims asserted against, such Contributor by reason
|
| 175 |
+
of your accepting any such warranty or additional liability.
|
| 176 |
+
|
| 177 |
+
END OF TERMS AND CONDITIONS
|
| 178 |
+
|
| 179 |
+
APPENDIX: How to apply the Apache License to your work.
|
| 180 |
+
|
| 181 |
+
To apply the Apache License to your work, attach the following
|
| 182 |
+
boilerplate notice, with the fields enclosed by brackets "[]"
|
| 183 |
+
replaced with your own identifying information. (Don't include
|
| 184 |
+
the brackets!) The text should be enclosed in the appropriate
|
| 185 |
+
comment syntax for the file format. We also recommend that a
|
| 186 |
+
file or class name and description of purpose be included on the
|
| 187 |
+
same "printed page" as the copyright notice for easier
|
| 188 |
+
identification within third-party archives.
|
| 189 |
+
|
| 190 |
+
Copyright 2026 Alibaba Cloud
|
| 191 |
+
|
| 192 |
+
Licensed under the Apache License, Version 2.0 (the "License");
|
| 193 |
+
you may not use this file except in compliance with the License.
|
| 194 |
+
You may obtain a copy of the License at
|
| 195 |
+
|
| 196 |
+
http://www.apache.org/licenses/LICENSE-2.0
|
| 197 |
+
|
| 198 |
+
Unless required by applicable law or agreed to in writing, software
|
| 199 |
+
distributed under the License is distributed on an "AS IS" BASIS,
|
| 200 |
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
| 201 |
+
See the License for the specific language governing permissions and
|
| 202 |
+
limitations under the License.
|
SHA256SUMS
ADDED
|
@@ -0,0 +1,695 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
ce4c490df05a10f0ed52e0a8a394f2c25cffcef2821be24007388020f6ff2726 .dockerignore
|
| 2 |
+
bbedc3fda3305820b977265f01b8619d87570a6739de3a5582c3464840f1e57a LICENSE
|
| 3 |
+
2cf529188995827d07413fcab3a2fda048506185376d9b439495365bed5d2a29 api_validate.py
|
| 4 |
+
556448845c2df6d212db03c121252d342f881a2d8db17f29a8d961b3a824cda3 build_native_checkpoint.py
|
| 5 |
+
ca51deb3c0f6b096bea1ff2312b1bab4a4a8fa083ba0e03b282ec4f8ef303779 calibration/importance.npz
|
| 6 |
+
c65c34c37633381a7322ff433a292f9e615d414bb834d1951b243ee16ecb310e calibration/instruction-calibration.jsonl
|
| 7 |
+
c13846b0b5a15946da39f98e49e0d6f0289bc30c46b646880e5f73ed17c90ffd calibration/metadata.json
|
| 8 |
+
bbedc3fda3305820b977265f01b8619d87570a6739de3a5582c3464840f1e57a checkpoint/LICENSE
|
| 9 |
+
a4aee8afcf2e0711942cf848899be66016f8d14a889ff9ede07bca099c28f715 checkpoint/chat_template.jinja
|
| 10 |
+
d0883072e01861ed0b2d47be3c16c36a8e81c224c7ffaa310c6558fb3f932b05 checkpoint/config.json
|
| 11 |
+
777d1d74f79a46f56f182d839b16f965e0b25e78b58e5ccc64fad9a9264f1f51 checkpoint/equivalence-mtp.json
|
| 12 |
+
7b24b0f45aa284c37a5038a1716d266dd2d25c1eb49b9dcf2ae0a324740b7f14 checkpoint/equivalence.json
|
| 13 |
+
a9d356d7bdf1ef4949e3e748e95b8e10ad9d4e2e838eddc38a0a7b6b94d1db8d checkpoint/merges.txt
|
| 14 |
+
25e6c97b125e3fc969f9ea99290e1d3ca43db93720b699eb46f21224d8dd3c9d checkpoint/native_checkpoint.py
|
| 15 |
+
34f774cf110e3ff94726171b52f6d462cb9b36651295b941efa85098ddb96ca2 checkpoint/native_manifest.json
|
| 16 |
+
27225450ac9c6529872ee1924fcb0962ff5634834f817040f444118116f4e516 checkpoint/preprocessor_config.json
|
| 17 |
+
556448845c2df6d212db03c121252d342f881a2d8db17f29a8d961b3a824cda3 checkpoint/provenance/build_native_checkpoint.py
|
| 18 |
+
c13846b0b5a15946da39f98e49e0d6f0289bc30c46b646880e5f73ed17c90ffd checkpoint/provenance/importance-metadata.json
|
| 19 |
+
ca51deb3c0f6b096bea1ff2312b1bab4a4a8fa083ba0e03b282ec4f8ef303779 checkpoint/provenance/importance.npz
|
| 20 |
+
26d3539b516be613f39563617cb9d33b3f83d401298125be392c80cefb8f7fe5 checkpoint/provenance/original.safetensors.index.json
|
| 21 |
+
59d35218b4780c71e90f2a7cf148cc1f8a0db44a4a5c5dbd90cda624d6b54b1a checkpoint/provenance/overrides.json
|
| 22 |
+
f15011f4b91e82c8b007acdf2c9c229954c2553a44c6dd3504010c07068dad43 checkpoint/provenance/precision-plan.json
|
| 23 |
+
2939042f81d63dd6d90b1b13574c2a505623a07c1684bb9e3d048182c25eed59 checkpoint/provenance/tt_eval.py
|
| 24 |
+
d7ad88592fb7de07d87fa0d10f9d1bd0d4dcc92856438fffc394f90e29e650d2 checkpoint/provenance/weight_mapping.py
|
| 25 |
+
88e0efffbbd322dc891ebc8cd891e6ff2fc96da16c5fde334b2b92e457c2db26 checkpoint/quantization-error.json
|
| 26 |
+
0d4f553fe416591e5d9dfdf45150f82b4d88b4fea590983b65fe126bf602ee5a checkpoint/tensors/00000.tensorbin
|
| 27 |
+
38bb1a27e20c64787cd377c3f18bf2117aa4ed8251d247aedb146bb4833b0573 checkpoint/tensors/00001.safetensors
|
| 28 |
+
b55492af5a288f52ebe592e39e2e75f6949706869b679621b41597619f9a7512 checkpoint/tensors/00002.tensorbin
|
| 29 |
+
817739482fde7bf45eab7b3d68831fb27490390928c60966d48084fae7127b13 checkpoint/tensors/00003.tensorbin
|
| 30 |
+
d0552d5ca5f484aa8fce00132cef93c03ebfdb2c0cf2bb79c0e48ca2d787f971 checkpoint/tensors/00004.tensorbin
|
| 31 |
+
b5a8cbeb04c20ae79baf8ccd161ec5a05403ee15c6974e657a883f98769b0888 checkpoint/tensors/00005.tensorbin
|
| 32 |
+
2663e15be67b452fcb6836f53efb57b1ebf7bc078889ef94c3d10973164647b6 checkpoint/tensors/00006.tensorbin
|
| 33 |
+
8d1d32db0f18d850c2af69f9972a86dc53b789f604cbb38e934078ba4e98e833 checkpoint/tensors/00007.tensorbin
|
| 34 |
+
dbceb9cee219c05d5331cca9d3fe128834dba6ac763db5216ea0cfc58d4c3e11 checkpoint/tensors/00008.tensorbin
|
| 35 |
+
1adcc930caac883061d6b927717fbc139896ee40754bc7970616a4dfc85666cc checkpoint/tensors/00009.tensorbin
|
| 36 |
+
1f0b6775ab6d1429b849a7547fa1e83d30d26a179fea984875d9845838208052 checkpoint/tensors/00010.tensorbin
|
| 37 |
+
752d18b915e7e3042dccce83624c75fa99879d390689a91b5fcdb72ff51ba8b3 checkpoint/tensors/00011.tensorbin
|
| 38 |
+
b5a07da9f758a7bd30153ee609e82be88ec4f6ec82c5693c72a66dd216315269 checkpoint/tensors/00012.tensorbin
|
| 39 |
+
e019672df4f09a8f8cbaa8851a9c0e6405dc243f8fd76f9ae64599bc0f6cf6ce checkpoint/tensors/00013.tensorbin
|
| 40 |
+
470eff8f523155457cdeb861853b0ef8c0909399f3922f70424dd60f35d39cf0 checkpoint/tensors/00014.tensorbin
|
| 41 |
+
0078928a85361a6b3b372725b25e9edb2332effce2dae748ad394819d624caac checkpoint/tensors/00015.tensorbin
|
| 42 |
+
88c0a18c5c2ea0c5bbaa99d9d3dd782b28cc184a3b4528adcbf66a0f27d82709 checkpoint/tensors/00016.tensorbin
|
| 43 |
+
161db47ca7a7747ec28c78a82b0b57ce46478ec7cfe2cd9f1f179fd0e44041b1 checkpoint/tensors/00017.tensorbin
|
| 44 |
+
4c2de1c8bc2518ffaceb0b3ed5c0a1ff1de0ffb3cd8b3d684635d01cafd96fd3 checkpoint/tensors/00018.tensorbin
|
| 45 |
+
6710c4b861036d61791c017dd22e7ec747abe741412e7f31d56b56014dfc2746 checkpoint/tensors/00019.tensorbin
|
| 46 |
+
37726ca2810b3114267b599c1fff6dd17c18b534be98349851a74a0031511fd9 checkpoint/tensors/00020.tensorbin
|
| 47 |
+
620b7f5bf138cef6f091b94e7bb4260e4e9693abcfc9257e6a7788af062a7ae0 checkpoint/tensors/00021.tensorbin
|
| 48 |
+
70cd2c36ae9b0c518ec4ed9ccba038448c8b07a77fc40a61dba1b4e47ed403b6 checkpoint/tensors/00022.tensorbin
|
| 49 |
+
0b176ee7254f919e4a8f4ce69d751888e7c0f91c3b23c1e828a04806741a4838 checkpoint/tensors/00023.tensorbin
|
| 50 |
+
d79eaffc8ef17f28f5f207c4755177f88a8b516bb0d5d66e3509a1740069801f checkpoint/tensors/00024.tensorbin
|
| 51 |
+
fd32920bfbff8c6c3a72a01da2988a8818c2633516697fa190b5918b748df0a0 checkpoint/tensors/00025.tensorbin
|
| 52 |
+
481c0f7d3572873b540465a618c3532ffd50e2a991593514e5e32b816698f1be checkpoint/tensors/00026.tensorbin
|
| 53 |
+
2a3d10f4d87bb8fb3b2f72cf68f3842f12fdc9a56e89557638877ede50ebe84d checkpoint/tensors/00027.tensorbin
|
| 54 |
+
a8dad6e14de206600af064221d3df2ed29a5223b2386babc9ad3d9c2d64e68d8 checkpoint/tensors/00028.tensorbin
|
| 55 |
+
f7fabe4cdca9624aa92fc3cfa0cea32cb0519d96c41966486b0dac8eb25b7729 checkpoint/tensors/00029.tensorbin
|
| 56 |
+
20e836010896fbde4c9d6a914a38349cccfb1957dccdf10bbfc471d5cbb6c3c1 checkpoint/tensors/00030.tensorbin
|
| 57 |
+
bccb4914324b470e6105d0658372cbe21022e1c7330438ec56eabc7f4fbc1341 checkpoint/tensors/00031.tensorbin
|
| 58 |
+
7f9285e2a7bec6d721ee5eb7592622dcf264d617b6ec28e4b98cfcb14f436b37 checkpoint/tensors/00032.tensorbin
|
| 59 |
+
58129242aaf0c9f119685a236316e6b7619fccfd2d8c359221ccf952db487159 checkpoint/tensors/00033.tensorbin
|
| 60 |
+
0abc905a66bf00ea9d014253c34f3f5a4881e65ebcb31c60d83cd8b80db64a3a checkpoint/tensors/00034.tensorbin
|
| 61 |
+
0e09ef1810508b030d88d75247116b302bf3d33deeb0b3f9dce7cc370e77bae0 checkpoint/tensors/00035.tensorbin
|
| 62 |
+
0050035ce84e831865a7de86f43cd1e148f08cb62305f5f5b59be2094b75eafe checkpoint/tensors/00036.tensorbin
|
| 63 |
+
2f3bdffe838c66cf2281b330b8e422260743f032b90b386413ae0fd555d919de checkpoint/tensors/00037.tensorbin
|
| 64 |
+
8fdbd257779bf19f48fb0e935eba3760b561cbbc8c0b0bea33709417eef07cf2 checkpoint/tensors/00038.tensorbin
|
| 65 |
+
34334f82bb0bb9edf4f47bbf1cabcdfdf2f774c130aa33f00c0f247cd605d90b checkpoint/tensors/00039.tensorbin
|
| 66 |
+
8adf3f60671f39e598c327d4e616b993a9191e7c126f639290dd203e6e057f0a checkpoint/tensors/00040.tensorbin
|
| 67 |
+
2484b92f54dddd557997a230280d8e2964dd867c5190f58138b23146676519ce checkpoint/tensors/00041.tensorbin
|
| 68 |
+
dc228baee2344db89aceffda449b0a3cfbec39f4a17436530a275544dd158fd3 checkpoint/tensors/00042.tensorbin
|
| 69 |
+
029abf2e4dc861d3bdfa61ec615f4e9c7615597e53de51d840a2b5b53cc88d7e checkpoint/tensors/00043.tensorbin
|
| 70 |
+
ff21a01ab0112788ff5bc8e6236180cc40fc12df4c2d27204800847d09bdd41c checkpoint/tensors/00044.tensorbin
|
| 71 |
+
e2e211fb1b163448dcd7de68460da8702e24fa27a19510dd6513310e360f2e60 checkpoint/tensors/00045.tensorbin
|
| 72 |
+
ab3d1fc4a9368572d01267ce3003df30a83be8df6ffa78251678636e92b3685f checkpoint/tensors/00046.tensorbin
|
| 73 |
+
dfd1e3c9738da0e46de25cd9cbc649cc0b8c9ecfe753d63a69597c3c010f69b7 checkpoint/tensors/00047.tensorbin
|
| 74 |
+
774604c9bae0aa6ee0cf89ab02f38d74c5deef275b7dacb58db10ebd7ece53f8 checkpoint/tensors/00048.tensorbin
|
| 75 |
+
c07f37d22d38e04cede1f49e6247955673004553187713014cb6e54f2319ab52 checkpoint/tensors/00049.tensorbin
|
| 76 |
+
20c083b06c60088c0288394470e1a7425f65304d926176553bc4aaa3e60411df checkpoint/tensors/00050.tensorbin
|
| 77 |
+
30752bf40eb40e8d815e402b5c89484688a0a1d364177ea4471f638ad8887d3b checkpoint/tensors/00051.tensorbin
|
| 78 |
+
4ce04d9b7b4cc2a72243657add46be6e17c3d2272ec72c89d7f9960cd8308bf4 checkpoint/tensors/00052.tensorbin
|
| 79 |
+
d0625ba5bec767267e8cf0f22d4b6b89662372643875ae8e6a0af2e186710e9e checkpoint/tensors/00053.tensorbin
|
| 80 |
+
8115106f6f2518fbc801e77bf13cbf41e7b28aaebbaf8d0cf530e27a8dcd196a checkpoint/tensors/00054.tensorbin
|
| 81 |
+
c2e5786d5d2d7efff55c2911d80386fef51d83f34f67fde75959afb1e4690897 checkpoint/tensors/00055.tensorbin
|
| 82 |
+
4b7f5fc94e230345e7effce8d0752929ff0d66ef175af15fcdd690c977422c4d checkpoint/tensors/00056.tensorbin
|
| 83 |
+
aa2bef86004685b25b45d7945c5100cefe6ba88ad0a6a317218f16345e73705d checkpoint/tensors/00057.tensorbin
|
| 84 |
+
6c2c261a1abf1fdd09a87da8c71d57241853d2764ac3d13ae1cf00a9e7b68f3b checkpoint/tensors/00058.tensorbin
|
| 85 |
+
298566b8f0f3c76e6a71ae4ffa04daa958f906eac3701983534ea53b5419afe3 checkpoint/tensors/00059.tensorbin
|
| 86 |
+
815db429f97496e3cf32b2632412fa465b1f901833d8ac5ffa9aec0a63f89766 checkpoint/tensors/00060.tensorbin
|
| 87 |
+
2fa0034ddff451981649961993d5cefccde251607dd1f05435148b8677688fde checkpoint/tensors/00061.tensorbin
|
| 88 |
+
6501754a904e6f5dfb10968ac855d94e790c96cd81c2684ac4d9304337389cf3 checkpoint/tensors/00062.tensorbin
|
| 89 |
+
dd793491aaf595f9d49cc54111ddbb000fdfaec4fe3cf7f006a1a82a92360743 checkpoint/tensors/00063.tensorbin
|
| 90 |
+
f2b848ca24564c87e0f5db270127764cedb97b689e4b1b9a9eceb3a775b48559 checkpoint/tensors/00064.safetensors
|
| 91 |
+
9ea198414c1d79d389ac94e866b8d084e55f7b241d9312b342477b6ab9fbbb7e checkpoint/tensors/00065.safetensors
|
| 92 |
+
504db46d0b60e1eb0ee028836486f9efdca5fa6a81c2a380af8cd435245c4606 checkpoint/tensors/00066.safetensors
|
| 93 |
+
d805110f65f3b87b3bdc1cf43133c2f89d5788c956d4d38bffb67a958c08835d checkpoint/tensors/00067.tensorbin
|
| 94 |
+
55248631595a532974415f42ae9ed0b5527690295e80e05aff638b47f013813b checkpoint/tensors/00068.tensorbin
|
| 95 |
+
4a1b9108344f1177e55cb08f3beabf9bd67f5b0e99731f2e9cd3728873ddecd3 checkpoint/tensors/00069.tensorbin
|
| 96 |
+
cdbc77f06c9c92f4d0c9750ca333ff8693127f3a83a6e1af5e61e7994507e236 checkpoint/tensors/00070.tensorbin
|
| 97 |
+
627840aaa440b20cf90441dbb13a359bc197f0798a81a4d9aeec305427b726e4 checkpoint/tensors/00071.tensorbin
|
| 98 |
+
9a76750c0876731a91ac5b068505327f6eb724917096e10166dc0b1e7e808904 checkpoint/tensors/00072.tensorbin
|
| 99 |
+
30b5c62ae765a8328a691832e3e6b786e115f52003290c1abad09b3a3ca8b138 checkpoint/tensors/00073.tensorbin
|
| 100 |
+
96ddc1961df1ff9ace0c3dfffd64299ae781fdaa997f71e7712f2c0b4516d07b checkpoint/tensors/00074.tensorbin
|
| 101 |
+
8f73151c88d081142422b03a198b667068198594460aa6e468316aef95ff1eae checkpoint/tensors/00075.tensorbin
|
| 102 |
+
d450d020d92fc8ca67f2711ae99fb3300fa50eafc9257ad919810127381d5a24 checkpoint/tensors/00076.tensorbin
|
| 103 |
+
8ee0c647eb1079bde11a646338cf3a49543fb21e4e89d626a18344e634f616f1 checkpoint/tensors/00077.tensorbin
|
| 104 |
+
8d7059cf1405f91c0b1a6473233404a91d4b5e6c1bb7a74c5caacc7b043cc4c3 checkpoint/tensors/00078.tensorbin
|
| 105 |
+
be95409ce6495807a69b30d23fa1443cf6f9efbe172fd0aed7c6c8f7a133c18b checkpoint/tensors/00079.tensorbin
|
| 106 |
+
6768a39ecf6b7f1640860c0516c81bfc7efbddbb00a1d1ff4fe1de2ba279f3c4 checkpoint/tensors/00080.tensorbin
|
| 107 |
+
8932cfd0e0cee9cfa7f5fbb76017df54ee96998630a86410b8825bc54ee7c1d8 checkpoint/tensors/00081.tensorbin
|
| 108 |
+
deb12edf58aba8544fe6075f9225606ed359ffa387a14467b0878ae0688763ec checkpoint/tensors/00082.tensorbin
|
| 109 |
+
def6b43065c2c20079e25159a11a4f083b7d1d4e9cb96b83a936de619a2775ba checkpoint/tensors/00083.tensorbin
|
| 110 |
+
e7cb7d7b6d214c9642826755baf01b2649bcb387cfd1a0ba901c852a39046792 checkpoint/tensors/00084.tensorbin
|
| 111 |
+
43f748a9525b8a24906ea38929a9ea5b996988d939d1409dc73306e116b58a30 checkpoint/tensors/00085.tensorbin
|
| 112 |
+
6feea622a98ff84898af115974c892bbad4f80ef9f3f97a27121a63a9335bb4d checkpoint/tensors/00086.tensorbin
|
| 113 |
+
838b0ed15bb1bf815be45340f1eda533941a14c388f2b204429eea3f73ca1fdf checkpoint/tensors/00087.tensorbin
|
| 114 |
+
020082276e7c5908ddb384880c007300a8f9e28e1125ebe928eda82f1832c435 checkpoint/tensors/00088.tensorbin
|
| 115 |
+
8cc775394ba1c159995ee3ea60efe889016b924dfaaf3aa42129215cc571f58b checkpoint/tensors/00089.tensorbin
|
| 116 |
+
230b248bec7296186b53f394d27eaf60909ba2f341049fa2f86369b91e33eb73 checkpoint/tensors/00090.tensorbin
|
| 117 |
+
a1f94dff13f4222f43672b88a1e97e94746c58637cbe18f8adb6e72a4e688bb4 checkpoint/tensors/00091.tensorbin
|
| 118 |
+
06026d439572908535d0f9c9b7995c3173a4ead4607254940c196aaa021d98a9 checkpoint/tensors/00092.tensorbin
|
| 119 |
+
e0852f6849ea78db8dd4b47f3b6afba80a55f8d07045bde68395320000387457 checkpoint/tensors/00093.tensorbin
|
| 120 |
+
6805413de6585faccb6a7dd7771652a5620b2d086a839e144b1400aa96b36a7e checkpoint/tensors/00094.tensorbin
|
| 121 |
+
d3d7293fb8a8ddfc653296fa1d2ff32b734234b26b397b939762d566d6072b81 checkpoint/tensors/00095.tensorbin
|
| 122 |
+
aaac2251885c04377ce4b6ab189b60539d857bcea37223341b241ed3512a098d checkpoint/tensors/00096.tensorbin
|
| 123 |
+
065e5c5806e20ea0c9d631d660ecfc730c00a0d3cd1a9953cde256c6fc471941 checkpoint/tensors/00097.tensorbin
|
| 124 |
+
4948c273dc7d8c589f6e0977cf9a7539b038b032cf911fcd4d1c361e71cb264a checkpoint/tensors/00098.tensorbin
|
| 125 |
+
1aeffaba0b7b684c80c4f93dcc1b51ce71fb78bff90f0d69573c88697128682b checkpoint/tensors/00099.tensorbin
|
| 126 |
+
6d636430d8665724ec6759fba20f599918856ff09b0e0cebeac1fea4b7da1fa8 checkpoint/tensors/00100.tensorbin
|
| 127 |
+
5eeaa4ec50442bc57c7a45e24e0901d9186d017f1ef3e2a8b7a217091d380c35 checkpoint/tensors/00101.tensorbin
|
| 128 |
+
8150f52f2c1b7c537caa71118e1f4a7eaf9c2413e2b3d02ac0413f92ee2f4d85 checkpoint/tensors/00102.tensorbin
|
| 129 |
+
dd530feed65d537c88b8b0664139a0a95c725eee19770a9b3641aa111000d65b checkpoint/tensors/00103.tensorbin
|
| 130 |
+
c973c181d1b9133eed39360be145c6d566efc5759bf8f2ad9380be15512a07dd checkpoint/tensors/00104.tensorbin
|
| 131 |
+
f034e2f598e8c8dcc938fcff3a6d5ea8e20c4580d6185e02bacd1e2f6f999e32 checkpoint/tensors/00105.tensorbin
|
| 132 |
+
4f8f5cab8e4423e70c29f321ead60ec9edf7933f987edbbee435d8f4a26cdec0 checkpoint/tensors/00106.tensorbin
|
| 133 |
+
6296caeb4c1fa89e462e9a615bfb647472dd4a7c4a3a4e5ac7d639648b142ac5 checkpoint/tensors/00107.tensorbin
|
| 134 |
+
9b62d517058c99f3deb5d729cfbe3613f9feb207f798b6f966a6beb4b87060c7 checkpoint/tensors/00108.tensorbin
|
| 135 |
+
4d1465a913340a25cbc0313d400da8fc8a04c4357a7a3bc2d35b812bcd27d045 checkpoint/tensors/00109.tensorbin
|
| 136 |
+
4b286183f512a413db858379029f128b605a3097a06af39d579b5534cd08ee96 checkpoint/tensors/00110.tensorbin
|
| 137 |
+
525a45f7dd52724d6cfa6503f31f1d0844a4f145bc6bc5296868293afe9f8f5a checkpoint/tensors/00111.tensorbin
|
| 138 |
+
b523d775a4af93ed9dcfa68086eabe191503fda8e3de9d29cada1dce76e9707d checkpoint/tensors/00112.tensorbin
|
| 139 |
+
a653a925df8698cfddcab95f192652ea8a0dbbb79a971c729e12e6880b3c41b2 checkpoint/tensors/00113.tensorbin
|
| 140 |
+
d914477f457833f972cd9cc02aa660b9ec75033aa8bef0ba7bea6356b671dfd4 checkpoint/tensors/00114.tensorbin
|
| 141 |
+
a71cc72015f9e1542109eae000ae5f9703ee01231aecdde233f1fbf07cfacb71 checkpoint/tensors/00115.tensorbin
|
| 142 |
+
92b7f0ead412df2684b2e2b127d4b94aad9509dbf37e1bc7563db47720c8b5db checkpoint/tensors/00116.tensorbin
|
| 143 |
+
66e7bd07ebb511bd69367a2e95a67545aee9c9aeb3b53dec095a4d2557fe7d53 checkpoint/tensors/00117.tensorbin
|
| 144 |
+
1c29e5b0e41911cfc94ea54ead6560dc9bea9f23132a88dcaab1288f65545d4b checkpoint/tensors/00118.tensorbin
|
| 145 |
+
9b6d8d7fcf4feebab07d931cc3ee821a2c31a305bff40d024e69d48e6724062a checkpoint/tensors/00119.tensorbin
|
| 146 |
+
29c5dc7fca5333477cd5d429e320ada19d51bcf8432f0579075b2041f1e2155c checkpoint/tensors/00120.tensorbin
|
| 147 |
+
f5eeffb974a6d35cf1e9a77e9e437ba2a3322b4f56b45aa27209c1cf0a06013c checkpoint/tensors/00121.tensorbin
|
| 148 |
+
1caa312f4302813480e8221e70bf1076ea6785d40afdc6066c7d6eb17d53d064 checkpoint/tensors/00122.tensorbin
|
| 149 |
+
fc6cb9d6de920085d034fda3afdff25a91c7746c838acbda609b7d0d5168b3df checkpoint/tensors/00123.tensorbin
|
| 150 |
+
143e45c903df39db2ae0da6e62319024cf7b477da298805262d45fec4957bb10 checkpoint/tensors/00124.tensorbin
|
| 151 |
+
49692a4fdfd572aa5b1e13019ca39f2797736b7ffcb3d0df1150f6a76e892e65 checkpoint/tensors/00125.tensorbin
|
| 152 |
+
7aa0d11b9445a76f873bccc254682230184c617e3bb86273f0db52f37bc5ce85 checkpoint/tensors/00126.tensorbin
|
| 153 |
+
ec2b07dc9222bcb54aa14feaa98ac7b26fda9c1d0a9cbd6c4e79b85b241b09b3 checkpoint/tensors/00127.tensorbin
|
| 154 |
+
244e0ecaa9599825163f1787480228ed4be81ea52ac52b70af161a0750f7013f checkpoint/tensors/00128.safetensors
|
| 155 |
+
8cb9a4c329d19a4c5763afe8d4327a5f95be8117fad21b59a2dff6f1f68117c9 checkpoint/tensors/00129.safetensors
|
| 156 |
+
6eb0e95a258543c55e6d6cff7350c8cd78396046ca778c5187901d3e84addde3 checkpoint/tensors/00130.safetensors
|
| 157 |
+
c79e2cb6b9bfa7b83cd08ca063a17af8ddb9749cf7c631f91873b7ffc6107b27 checkpoint/tensors/00131.safetensors
|
| 158 |
+
869a8cf892a761640fe43c26ff83b818647ef99a69f4b4dd4762c4e441552e74 checkpoint/tensors/00132.safetensors
|
| 159 |
+
34d57d4fc71b4f4c3e8ba820976fcfe321ec1cc48356097023bd6bf60a7ce88b checkpoint/tensors/00133.safetensors
|
| 160 |
+
157df1bcc23a9e7b668929bc7abd40487065370399f83d9b62c97c58d1c373aa checkpoint/tensors/00134.safetensors
|
| 161 |
+
48190215f980ca8a2f125691922f2aff80f80e55d13c68c96e94e99e3c5ec42b checkpoint/tensors/00135.safetensors
|
| 162 |
+
85f0c3f3ea24c9dd04b5a4ce01ed8c8f0ba6f298b274da7d5f1b1f0bb2611fcf checkpoint/tensors/00136.tensorbin
|
| 163 |
+
29d88cb438fdee0ae26ab244fca55c6b3c7100ec94e0277a45ba5bc64c732b1c checkpoint/tensors/00137.tensorbin
|
| 164 |
+
d8e5ebfd1d78ca5d8658e0d8b1229197fb9d5223de252684e24cf4aa47b20be5 checkpoint/tensors/00138.tensorbin
|
| 165 |
+
50ca52b92ce1b395c3829b1b8146ab8a5eb575ea7cded4ec9ca2eee22a45f659 checkpoint/tensors/00139.safetensors
|
| 166 |
+
2d3034a7e1059dc969cf4e5b2e192a63cbc14cd97f25c915650018274de9fcd6 checkpoint/tensors/00140.tensorbin
|
| 167 |
+
f7fd91d22fc075ba0e6b9609fad3087cb1f09b13ff62c69f5397145798394bb5 checkpoint/tensors/00141.safetensors
|
| 168 |
+
1600211307e4bc218c89e5db4a32d344e21dc715f8a3a20d766b211cb391c901 checkpoint/tensors/00142.safetensors
|
| 169 |
+
995187205d327324d3af45d009e2ad8abc13eacd7ae87069b15ff9302951c63f checkpoint/tensors/00143.safetensors
|
| 170 |
+
c32b964236c31808ad77be52d2b1ff21ee0c25958c850c598a4dff76f83d5d9f checkpoint/tensors/00144.safetensors
|
| 171 |
+
8ad4b56fbc765b805bdb0862549ec986c5524280c17904aa3c26056081795abd checkpoint/tensors/00145.safetensors
|
| 172 |
+
902b2a8472d969f2474c29fc1eda2d68b7c57715b3e2465502d1da04d681a8e6 checkpoint/tensors/00146.safetensors
|
| 173 |
+
cf613575983164c6ce667256f6e7b1ff875029081e25d0b0ddcc1429761adb3e checkpoint/tensors/00147.safetensors
|
| 174 |
+
553a6b16596276b5e8f7e2af1f37202f6c29b47d68ff8c59ab50d9c825af8481 checkpoint/tensors/00148.tensorbin
|
| 175 |
+
d4f49cd50a3087c2a3e5919bd41faf17e5ad0a0fc4a1d5f35f1ab8dd56c3b53c checkpoint/tensors/00149.tensorbin
|
| 176 |
+
e3196bad4a922745fa7fa606d082eecfc0b2ab0fda2788b948d17ae3f4188084 checkpoint/tensors/00150.tensorbin
|
| 177 |
+
a299c93deae9d2f4c966f90c6692837b0508d6b05b98d84e8b426005e811e570 checkpoint/tensors/00151.safetensors
|
| 178 |
+
378d28e5271c53993c7666873535a219f0adb6b9730ea67d718d236a36219834 checkpoint/tensors/00152.tensorbin
|
| 179 |
+
bd93ecf355364f054da40a0af6908339ccb2ed0e371bb0a9bfc2823430905b49 checkpoint/tensors/00153.safetensors
|
| 180 |
+
71d5f229c2e437af23843635cdb07ebb2b9c27d370c6c662e0a9100f6c385008 checkpoint/tensors/00154.safetensors
|
| 181 |
+
b3c0f64c8b66a65010b61ab9cb50d6b797169df3c310178c68a3f0f1a43b0df2 checkpoint/tensors/00155.safetensors
|
| 182 |
+
000d0e3f7dcd34472148e6e7adfaad8d8412b30d58483415c3c51616e1f22678 checkpoint/tensors/00156.safetensors
|
| 183 |
+
e28ac8299e73773142f1dd63f860cc428b811d420c279f11c34e23ad36cdbe66 checkpoint/tensors/00157.safetensors
|
| 184 |
+
228ea2f75360ded7a5de13075f6f23102ae8ac8649b83c92214dd76f9b95c407 checkpoint/tensors/00158.safetensors
|
| 185 |
+
d9f5e04052e51db1a88070ada59571ff488f875ef127f47b661878978b7e81a8 checkpoint/tensors/00159.safetensors
|
| 186 |
+
ecacbc5c120ebbc4ebad237a0bd2cc1108493a3987f455c0e93b1e2f17804a09 checkpoint/tensors/00160.tensorbin
|
| 187 |
+
9a15b402778a2da7e01bb34d4621ebb284a5396ffd45667891da720a417ae939 checkpoint/tensors/00161.tensorbin
|
| 188 |
+
fdded81ff117ea33891a4880dcb1ab95f3f3df946d1ea1858c0e507214e4a2c5 checkpoint/tensors/00162.tensorbin
|
| 189 |
+
045109ea1decc75d5396eadd9513d235f2999b16ad2c97dce7ebd01d5e20ccbd checkpoint/tensors/00163.safetensors
|
| 190 |
+
1a462260bedbdba19b804582501df638b813b70aa8d8e5000682c3132ee8c601 checkpoint/tensors/00164.tensorbin
|
| 191 |
+
60d223bbc31f50c5a48029c2d4040643891be2615fbe1fa9fd7b614a2e7946ff checkpoint/tensors/00165.safetensors
|
| 192 |
+
12f6be0d4e286fc1ac09846169bb43ce3adfe797f134989c74ae1e5aa23a17c2 checkpoint/tensors/00166.safetensors
|
| 193 |
+
08296143d53592bf5f003ec330ebb7cf4349b11efa3cdc0f0c3247d77ee63f39 checkpoint/tensors/00167.safetensors
|
| 194 |
+
fd38b38978f626938c265223ef4350c26ec7712335b9baa4b5e380e3a7f19e79 checkpoint/tensors/00168.safetensors
|
| 195 |
+
1b43c85b5e90e107df7b3ac19cb6c133d14c58dd290113cf3cc6cb6043a8175d checkpoint/tensors/00169.tensorbin
|
| 196 |
+
9478d76047a7d3d8b5c6cea92497bdc906d16168542ffd5796843518da1aea65 checkpoint/tensors/00170.tensorbin
|
| 197 |
+
47bf913af40449fe112b4d2e4aa59bfd0fb75e57d1fcd37facba51c321c74441 checkpoint/tensors/00171.safetensors
|
| 198 |
+
1cfa7262159e4c6525959d64975d4fdc1a50a378b37d7e8a5fd1d602690c1b29 checkpoint/tensors/00172.tensorbin
|
| 199 |
+
a782c4052a3e575d6300574b984aff59726ee1f1ec96d08e73e40f3b2e5acfa8 checkpoint/tensors/00173.safetensors
|
| 200 |
+
0b6f34383578b3105649d1f95c5173bdb88376c8d9ab32158cc6a6bb0bffd4a0 checkpoint/tensors/00174.safetensors
|
| 201 |
+
c6062e3d619185fbd9f97a41aeebc3c85f09589fef9381a8a76f88775526555f checkpoint/tensors/00175.safetensors
|
| 202 |
+
acb0e88a3f39981773996365a34edafa6d24b1b9ffa0fe1ac65802b4e98e30a6 checkpoint/tensors/00176.safetensors
|
| 203 |
+
e937c3bfc5ead2cd8933ced13c1763939e27519b4a5d44739c4ee382c8730041 checkpoint/tensors/00177.safetensors
|
| 204 |
+
f7fc5427d819ad5e42365571554e525ac72355ce61eb63033cf03adeb217b06d checkpoint/tensors/00178.safetensors
|
| 205 |
+
e2d9baf4dfa310085e1805120db2aea2a49e0ff04f8dbab5d2603a6950244678 checkpoint/tensors/00179.tensorbin
|
| 206 |
+
eef707a8e14ac8212bdcbae463986fcfba77bcd57efa8869734c3f42a88d9367 checkpoint/tensors/00180.tensorbin
|
| 207 |
+
25fa98dda4cde0fb27fa99f9f14dfd7bf52e7d381b702f92c7645fd1fb22b37f checkpoint/tensors/00181.tensorbin
|
| 208 |
+
d762db369d89d0483ad0dcf213763e78f3a065026d782dff2fda6a5bc4fe24f1 checkpoint/tensors/00182.tensorbin
|
| 209 |
+
98084788b206a21fb9fcc1694b0025784ceac47a77fa10747255d2b493efad16 checkpoint/tensors/00183.safetensors
|
| 210 |
+
839f3d059da405d2371d6f1388c1279439942e11624d8d9697d36c5f4b81532f checkpoint/tensors/00184.tensorbin
|
| 211 |
+
c7b82e9ad9d4baba91e69459793eac00b2fac5b86d6b6213afdbb873f09d3c6e checkpoint/tensors/00185.safetensors
|
| 212 |
+
5278825dbe62cdb7b38b364e5292e6131c4b741c4091d039adfa093b90616ed9 checkpoint/tensors/00186.safetensors
|
| 213 |
+
f5797fdecb3473fa94332bca67ef8f49953c77bcefa8b52e8d158622292d6efe checkpoint/tensors/00187.safetensors
|
| 214 |
+
e11fc1260bacf6ac2142b57926cd18cd701ad27c4cfc56d1c2e6f05578dae2fc checkpoint/tensors/00188.safetensors
|
| 215 |
+
166512ceeaf961d4247de5a5de9cc281541c1459cba9c2b34e923ecbbb7d1075 checkpoint/tensors/00189.safetensors
|
| 216 |
+
f6fad43254aa6bdc0ddfea48edc2bec12ac864f69d0ed29682440b24ba91f18a checkpoint/tensors/00190.safetensors
|
| 217 |
+
d4fe3fe437a16c955c1206621e12d3b80fae7632189448bf2531091d77c27a9f checkpoint/tensors/00191.safetensors
|
| 218 |
+
5f8d47f54e42a4c46ca3390449cf2b469c63545780b25c586a5839b5c5fae32a checkpoint/tensors/00192.tensorbin
|
| 219 |
+
ce532e59eb33e19c90173f5cdc273658d21b9f390ee6c5c20e7331e67ccb88d4 checkpoint/tensors/00193.tensorbin
|
| 220 |
+
b1344d5ac0f3d006c7263ebf8d2d22fc1c9a6f9478927198bf16fa91897ec318 checkpoint/tensors/00194.tensorbin
|
| 221 |
+
ffb9c4270c095e8c6f2f78a225b619f3bdabe6ad25ccb7dbdb76bb95130602cc checkpoint/tensors/00195.tensorbin
|
| 222 |
+
94f3b04694b595970690995658b46f2f3f8f90838a91731888f287971f76caa7 checkpoint/tensors/00196.safetensors
|
| 223 |
+
e9a63febcd18a3d38600329657bb6e2fbee3e18a4d400c1f1385fd04ee5c58e6 checkpoint/tensors/00197.tensorbin
|
| 224 |
+
5d9bc051681a0f93572e1e462964a2681091d591236c24692a058b59f42ed098 checkpoint/tensors/00198.safetensors
|
| 225 |
+
bd4af1de2e57ee4ef1ebfe29f8ad41850be4c3ad4880be954ce64eee55ffcf91 checkpoint/tensors/00199.safetensors
|
| 226 |
+
59d24ffc0517b3ebeceab7ea318aa24a70a7c6841a2c66b02eaea462446bc505 checkpoint/tensors/00200.safetensors
|
| 227 |
+
4ca069f655ed08c888b90c422f40e80c4e4228d603d186bcb82694288b977b48 checkpoint/tensors/00201.safetensors
|
| 228 |
+
fda9ec3799cf6262cd9eac9fd8356a04adae8425452e46833efc024bc05ca6c2 checkpoint/tensors/00202.safetensors
|
| 229 |
+
f338a70207e2ba55198c776b95992e187ab399cedd96cecd5ab357b5e0255175 checkpoint/tensors/00203.safetensors
|
| 230 |
+
0654336d63056bb427450ee26c9a40c7c6bfa356685c73ddd4621b67e0fea59e checkpoint/tensors/00204.safetensors
|
| 231 |
+
fd7b879af6737c6b16faa652ecdfc38573d504abb0062af5ed090996c8b3fc50 checkpoint/tensors/00205.tensorbin
|
| 232 |
+
d00c56820f695d40380fa2374ca98b5f7dd097f7a79b437544db69e213d42a50 checkpoint/tensors/00206.tensorbin
|
| 233 |
+
f2b245ca51cd474595c982928b0488b37b40596a6827cd1681c3709e4f7428cb checkpoint/tensors/00207.tensorbin
|
| 234 |
+
cd3da56d5e52804bc958bfeeecf5cba62f21266b37184e13d735ff30573eb778 checkpoint/tensors/00208.safetensors
|
| 235 |
+
9bf07a31dd7336993059401aa9b689c41db6cafa312accd31d1b25383062cd37 checkpoint/tensors/00209.tensorbin
|
| 236 |
+
c576ff0a8fc19406026cbae00a9b2fc291a8e9fa2346c02c3f95ec9132e34961 checkpoint/tensors/00210.safetensors
|
| 237 |
+
219f3544eb5c4a08a710e711c9d9bc9122dbec1f51ef72f4d2dc76b3888455c7 checkpoint/tensors/00211.safetensors
|
| 238 |
+
271c57142128391983d9b7e7011751e3b1d131c6d887e86b197560c53f3cb7e1 checkpoint/tensors/00212.safetensors
|
| 239 |
+
8cffff78ca9c90a578d1878a2d0fb46c00c38e908389f1d214c295a1d5540351 checkpoint/tensors/00213.safetensors
|
| 240 |
+
ebff2ddc9ef56e39f07fe446792989f4c2bcd1a19e843f83b7a1f64c2bdcb917 checkpoint/tensors/00214.tensorbin
|
| 241 |
+
ce9a6334631e85e1601ef178fe7a9eb981d42c92351a094a0f4175a8035a89c9 checkpoint/tensors/00215.tensorbin
|
| 242 |
+
c30087ff46aeb95581e20b3f1056b9ee8634052572d3e34205570db01869357c checkpoint/tensors/00216.safetensors
|
| 243 |
+
0921824fe3a5bf5e821bbdeb3b066f416806c155439503815721a9e8a20629dc checkpoint/tensors/00217.tensorbin
|
| 244 |
+
f4711cf8c8052221009cdb3346d8e8806ef6d17f471f5fc1ffd8be2b8181ceeb checkpoint/tensors/00218.safetensors
|
| 245 |
+
79c1760736b8a2e0bf7c50095c419d03f946fa2b7bf5d538c0cd3c11d0750cf2 checkpoint/tensors/00219.safetensors
|
| 246 |
+
ec7428e1738dd71fcc0970c485fe1d26d115815ddcc4e299c349d5fb9de7045b checkpoint/tensors/00220.safetensors
|
| 247 |
+
67ca646b743ed2b9647c66cc6185d23e731a29cf67c3e821f4ca35dd66882505 checkpoint/tensors/00221.safetensors
|
| 248 |
+
8354d9c03e3e5800211294eb49c878715c47b43a7693f0b296036867dfdcea56 checkpoint/tensors/00222.safetensors
|
| 249 |
+
f19c7ce89230882513b6e068004d1fa45e0dd492759575b4096c10f55522ed17 checkpoint/tensors/00223.safetensors
|
| 250 |
+
95b057d8b3514fd4eff05a9f9ebb5a7a85af464c706d4fe375b0e0c33ab81f42 checkpoint/tensors/00224.tensorbin
|
| 251 |
+
267e850dd6a6c555d6344aac1094250dd3a50ea3ca8dab38edc2038351fab54e checkpoint/tensors/00225.tensorbin
|
| 252 |
+
c2dd907e1fcf1cca8514f4da83532dc1ff9664c1bb3ea3de72dc45f41e801da3 checkpoint/tensors/00226.tensorbin
|
| 253 |
+
4ca188f46ca8251a3f489da651007cc83d4a9cfdfccf92457dda8ba848aa8aa7 checkpoint/tensors/00227.safetensors
|
| 254 |
+
5eb385487013d1e2d55cc95607feac7ea9d6363f43d7fcced207d92b6f98e11c checkpoint/tensors/00228.tensorbin
|
| 255 |
+
e9c3baa03bd6fb2d322a1730c41f23a2c84fe3f7637d466e4dd8b8781e0b9896 checkpoint/tensors/00229.safetensors
|
| 256 |
+
d1bc21df4e1a98d55d81c36805f901a1432d6a2358832c1f67ad8a3e4625c84b checkpoint/tensors/00230.safetensors
|
| 257 |
+
5cbcfc3e3055121805f6fe16945125ed7b1bdaa817c6cd6d4e5326f52fa3041a checkpoint/tensors/00231.safetensors
|
| 258 |
+
fa466d1f2f28c3feb99612085bf3d7564f3f419ea713816b4754430afb35bd00 checkpoint/tensors/00232.safetensors
|
| 259 |
+
5c5f72b121266139d80b380a74abfee045d2bb125f3a03a0827e3f17cce47588 checkpoint/tensors/00233.safetensors
|
| 260 |
+
c454016bb441aea8e5a4836bd02f524252e9d9a0929eec452c6373aedfd58b16 checkpoint/tensors/00234.safetensors
|
| 261 |
+
22bee2201bc52bcaaef8507de5fe8fd26a4fab30f59792f5dc81eda669f982e6 checkpoint/tensors/00235.safetensors
|
| 262 |
+
42dc7c17e02829dac860be23e0ab073977133274397da682c72412a4a5d9cad6 checkpoint/tensors/00236.tensorbin
|
| 263 |
+
056ac15c8eb24418276543a1016261dbc181e8b78e725261a21a0477d393f33c checkpoint/tensors/00237.tensorbin
|
| 264 |
+
4c71c8d1026e980410e81045a5aa01184c39a9902e4dedfd7243550bf6aa0dfa checkpoint/tensors/00238.tensorbin
|
| 265 |
+
b0d3d256640f80cdd3b473e587b786211d14a11d7accc0094d2427d2132528a4 checkpoint/tensors/00239.safetensors
|
| 266 |
+
d256c4586dd4b6ed5789798b8ee19215add1be50885227e7c4bca14fa80cb6dd checkpoint/tensors/00240.tensorbin
|
| 267 |
+
322d70565252e96f4181a9a3609f0b2cd99af77143def7fe9817c516c0a5f420 checkpoint/tensors/00241.safetensors
|
| 268 |
+
32911e732d3c2393a1e51fa830a6023a7a385728eb4d9316d04323c34325438b checkpoint/tensors/00242.safetensors
|
| 269 |
+
6b3c4fa2fe5802a96eefa1e2b1608acd70ab2aa2a845af4227750704fad8c07a checkpoint/tensors/00243.safetensors
|
| 270 |
+
51df336e61046f72f6b77db284c58a0fc1353378495c919e9b29885837796eb7 checkpoint/tensors/00244.safetensors
|
| 271 |
+
ed061db718e9602639671015229e95ac54cc989452ee482f4e4dd9f13aff7cd1 checkpoint/tensors/00245.safetensors
|
| 272 |
+
de6444dc0df6eab11621b682cef3420c769acd4e513334adffdeacbd7f5c8b61 checkpoint/tensors/00246.safetensors
|
| 273 |
+
2fa326e50aacde9f2baa954c6fc51bb1c3f562434538b4644d6a261c62243b1b checkpoint/tensors/00247.safetensors
|
| 274 |
+
90e0b8fdcd1521d12eaa8b57629033e2763e33694f91442778fd2953d080a0f4 checkpoint/tensors/00248.tensorbin
|
| 275 |
+
abd6bb00b46a52129d65a77f76b629fe501a9439e0dd539615489802b68f0458 checkpoint/tensors/00249.tensorbin
|
| 276 |
+
1637792f7773121b9d22743217bdcdc84d7805ae7d2ead037b06a878d59a8026 checkpoint/tensors/00250.tensorbin
|
| 277 |
+
51c94c47de55b877011bc1bbc9013755b077614684a339649e39d9a2079ff586 checkpoint/tensors/00251.safetensors
|
| 278 |
+
47d47160f3c2064b988723b88a43336d24d201009ef6fbd2b1533e005c513343 checkpoint/tensors/00252.tensorbin
|
| 279 |
+
2b38cc904ef321ce74781601406446209383646dfdbeaae5726dfb008eeafd0c checkpoint/tensors/00253.safetensors
|
| 280 |
+
38654ce7b79e1c05aa61dc9b34c4ac7b29fb06847ccfa9a36d4de4a6b2df3265 checkpoint/tensors/00254.safetensors
|
| 281 |
+
1c7bcdd21ab2089806a4c7b8bd94f1098e23005cb5f8baa20f2cb48a876a1bc3 checkpoint/tensors/00255.safetensors
|
| 282 |
+
2106b1e6bc57138bd7deb2cf6450ff5176ef6707046f3e1d37477fff69f1b73d checkpoint/tensors/00256.safetensors
|
| 283 |
+
48fbdeb1777bb5d9a7d1d8f1f37bbc280953d6ebb3cf3c858a86205663c34fde checkpoint/tensors/00257.tensorbin
|
| 284 |
+
57890eb05e76d93c95beef5106e17143ecbe3bd1984136c28c3df0b38a08619e checkpoint/tensors/00258.tensorbin
|
| 285 |
+
89448a527c766d504122efe57ed9f8834fee66fcec3b93ef1e85d93f2756c47c checkpoint/tensors/00259.safetensors
|
| 286 |
+
6157fee24d41d4c8d0fd5c4d85c59f61054ef36618511d18eba845f0d826e01d checkpoint/tensors/00260.tensorbin
|
| 287 |
+
a0060e8d2f168f6f6dbab9a65cab7b3dfcfcfc1b1d1429f1c5660cbd62907d49 checkpoint/tensors/00261.safetensors
|
| 288 |
+
df0f9398360b1edaa5a59a1df1f90a89924c4bffdf076594a876a01a1a387b2e checkpoint/tensors/00262.safetensors
|
| 289 |
+
473f681d01a901d0a45a60f1d70ed59e2655ddfc48f28dce62cfcaffc3e7d3bf checkpoint/tensors/00263.safetensors
|
| 290 |
+
c4352ee4b4830dded883be793cb385ea964f07e92c4e52ca57c35f13fea8f4ef checkpoint/tensors/00264.safetensors
|
| 291 |
+
a853fce2667fb0bab567fb770c58ca23aacdca51bfad1923103f0840c2812834 checkpoint/tensors/00265.safetensors
|
| 292 |
+
37ab29260417aa5f80b3a9886a3d4d7dac978ad78a16e43e3f48afc3515cd6c0 checkpoint/tensors/00266.safetensors
|
| 293 |
+
5822f43f164be88896a302d49a93372bb3ff29df60af1ca72ef7119d5eb8488c checkpoint/tensors/00267.tensorbin
|
| 294 |
+
b157e6416e4a72a0e2bb5cabdc1d390d04dba42b3d25fa63dd36ee53c59b6140 checkpoint/tensors/00268.tensorbin
|
| 295 |
+
665c065bb694dab4ac72c5ee6b3040dc2a2dbed3d368108eeafa41960e86faf6 checkpoint/tensors/00269.tensorbin
|
| 296 |
+
2e61d7969a51d475cd21fff4de9c213dfa81f1934c9d929285de3a771149fe41 checkpoint/tensors/00270.safetensors
|
| 297 |
+
8178aa75f45a4487085af2de907c640889616171c4abed815dade7b6e135d5f7 checkpoint/tensors/00271.tensorbin
|
| 298 |
+
8398bcff821b7e10702c53bf8997e405df7c8ee648c3c17186399003b4cabc51 checkpoint/tensors/00272.safetensors
|
| 299 |
+
6966a9e81d1f01b4d2176e2d41bbfb938a06ba1326993244ca4d71466ec3f3ef checkpoint/tensors/00273.safetensors
|
| 300 |
+
5957cb85115b700fecdbe670ad7b126c483abc60f877054fab0c09c43077bcf8 checkpoint/tensors/00274.safetensors
|
| 301 |
+
dabda53cca267185337d397d61cb1aab6c384d5665b990546dbe3011d38cc9c4 checkpoint/tensors/00275.safetensors
|
| 302 |
+
a0c25b21e3d249d526bad702f84b34f102b355578e0f49b31c8eebe3e04c09bd checkpoint/tensors/00276.safetensors
|
| 303 |
+
46d0a26891810a7ebb2f6b0be9b11c3ac2626d8b20a43ddb61753a847fd44c2e checkpoint/tensors/00277.safetensors
|
| 304 |
+
f425bccf2ed1ad258158a29258e86811e10b74acac215c44d1b3c19ac15b2218 checkpoint/tensors/00278.safetensors
|
| 305 |
+
aabdd93d47a5e14bbdb92558776bc0efdd1ae1a4e51295716347d596a1f6e61a checkpoint/tensors/00279.tensorbin
|
| 306 |
+
9f00a4543ccdc6dea4b3d27a5d0dee7e9f8a5bb76153de5b1e710e15d97877e6 checkpoint/tensors/00280.tensorbin
|
| 307 |
+
8d7beee32e834a01c4909aee73da34ca9cc788d2ed6587d0b056ab1bd3f7d715 checkpoint/tensors/00281.tensorbin
|
| 308 |
+
039ffc93499858c43e7679895ee183da223b815ac302eb239abf50a8878f03bd checkpoint/tensors/00282.safetensors
|
| 309 |
+
b85710e21c8c4d9e76c7c7dd2fd05acf8caf3e32a17095930c3f9d5afec97ad8 checkpoint/tensors/00283.tensorbin
|
| 310 |
+
7f25dd97556bd37d172034a49879985dda408b6c324b45ecb36764d50b12f28f checkpoint/tensors/00284.safetensors
|
| 311 |
+
7f70fe196dd9667d5c6dbbc39db294acf42f2c60d9abb8c7053dce8ba5036e3b checkpoint/tensors/00285.safetensors
|
| 312 |
+
1a9978ee692a87a4f74bc795fce8cca8d5629362201d1e36ae98159c76d4f505 checkpoint/tensors/00286.safetensors
|
| 313 |
+
01de794ee6fab9abd381f43b9576f35435e33a566ae2b43d13be23bf60f2e5c1 checkpoint/tensors/00287.safetensors
|
| 314 |
+
5a24fde5eb9a721434c43729db60c14c2ec1b801541eb3dd5264b75d2220e13c checkpoint/tensors/00288.safetensors
|
| 315 |
+
dbbb56a36f63e54c15485e48ca47ddf16e1faf0f1d5b4ee3ada3ed6e64495b32 checkpoint/tensors/00289.safetensors
|
| 316 |
+
3f51f688965ae2d583119e9535e40d65bddd33d10f4b2f88f7af84959cec9ff7 checkpoint/tensors/00290.safetensors
|
| 317 |
+
4a1eb64dd8ce24800bb597ce271cd2c77488222a453bde4da27cf196d62b2455 checkpoint/tensors/00291.tensorbin
|
| 318 |
+
098dd64c24ee931b2a77664e5da03b42ef5490fccadbce75b9545f9a77a49391 checkpoint/tensors/00292.tensorbin
|
| 319 |
+
9de9a8f1122f29ef9f84a024bb21290f951f77998f93973df6c7f194a6668ed7 checkpoint/tensors/00293.tensorbin
|
| 320 |
+
51c2b37f53a5877a828b1224a397e404c9f7365cfad2dcd9f3a14282b5852587 checkpoint/tensors/00294.tensorbin
|
| 321 |
+
3558af3c9aa5252e86e7e4976606a40201d5495b2b60ae73f6b7d8cd262c36ff checkpoint/tensors/00295.safetensors
|
| 322 |
+
0ff4091c31a64c8f77fa19072530b34e954205c7b90609f69498de734946b1f9 checkpoint/tensors/00296.tensorbin
|
| 323 |
+
fbf2be4572bcb98c6e59e0b0770cdf26d3edb7be4adff8e0bb1a8bdee3726c38 checkpoint/tensors/00297.safetensors
|
| 324 |
+
85d66979e3fd4359e5f0fba07d9354f4852a363d4ab373acde0ce79a2fc6ab8d checkpoint/tensors/00298.safetensors
|
| 325 |
+
9c439ebad6ef276ee882906b8a49ecb9f5ebdb2a14e96c350765f4086683b519 checkpoint/tensors/00299.safetensors
|
| 326 |
+
6497fe2be1c56d055ec8749f5a374cce098f391327006014e70e215e56b00d05 checkpoint/tensors/00300.safetensors
|
| 327 |
+
c9c72eec16aa6deb7eb8244dff9ed4c6446e0e3e5c67962452cdde1c9015f505 checkpoint/tensors/00301.safetensors
|
| 328 |
+
0d90035447f1ef89d84b3004fcae47ad7669d468547fc1c7c4e74a303a7ef63d checkpoint/tensors/00302.safetensors
|
| 329 |
+
50fe0d22f45354d5727dafc50e99aca18f4f77de8d97a5b2d15ebfc5e9f8cc60 checkpoint/tensors/00303.safetensors
|
| 330 |
+
cec4299aa26f9b2efef0f94e7e6c388dc15e2010b5426564234252b17dbe915c checkpoint/tensors/00304.tensorbin
|
| 331 |
+
978f9b48a6d92a570b65351156d327d8225d2d40fa67fcc6896dac31b935873c checkpoint/tensors/00305.tensorbin
|
| 332 |
+
2fd617ac3a25d25322264cd62f1d7ba9308f5d36815529760932e596d1cbb745 checkpoint/tensors/00306.tensorbin
|
| 333 |
+
560029f02c32b5d3fb3af13065cb546add72c4e87262dbc2380140eee989921a checkpoint/tensors/00307.safetensors
|
| 334 |
+
740b19bd5978a9fc3044ffda2b1625b7023e284fde2189e0b7159c5a581a4fa4 checkpoint/tensors/00308.tensorbin
|
| 335 |
+
1e4ad227794afea73281315e2007b9fbe815a125ad8fe2c059801536908dec5b checkpoint/tensors/00309.safetensors
|
| 336 |
+
ee3c99d0dd30cd0e96d04e714bf684c33d15914ee4b4256864b29ba317379b99 checkpoint/tensors/00310.safetensors
|
| 337 |
+
b5c9dc07eba68c36550ac9cc47493ddb1746287e9c497dd2acc528c05fa96de9 checkpoint/tensors/00311.safetensors
|
| 338 |
+
8cf4376df8e48b902830ffb21b4ee8052461e621892b2ebe6d27b1f6f7b281df checkpoint/tensors/00312.safetensors
|
| 339 |
+
17f74ce7eebed8d5234996aa6eb2af6770db305c0b0d829331afd342745c0826 checkpoint/tensors/00313.tensorbin
|
| 340 |
+
7138c9f183b12a3175fca61261a6ac5d174e0e03ce3688ea69e2fe4f342acf7e checkpoint/tensors/00314.tensorbin
|
| 341 |
+
795db17e6f5b1c3108fcb6dfe847eaa8121de78d58474b38b72e15b75bc569ab checkpoint/tensors/00315.safetensors
|
| 342 |
+
eb014f2fbaffd21916d2377855333643343f725bb218c53f60543cfca6b16847 checkpoint/tensors/00316.tensorbin
|
| 343 |
+
e312836237d70d7f099a6f1b4089b0c9b187b01bf6b91ba7ab438e4056e5c9bc checkpoint/tensors/00317.safetensors
|
| 344 |
+
2ecda668d3873790f22ab4576ef688ee7001c03201de31e8f894ac3903e1c526 checkpoint/tensors/00318.safetensors
|
| 345 |
+
fe5129a5eaf60df9d4ce13f4fbc10cda196a155d94407f7a6fc44f34e3afdf11 checkpoint/tensors/00319.safetensors
|
| 346 |
+
f6cca8e63072909f0de146423e1e7e7ae420e23f1540cb348bee2112d6103045 checkpoint/tensors/00320.safetensors
|
| 347 |
+
eb5131299b71e62d01e1c337dd7e94bf515195b23376afc581fa7a1d27e096e7 checkpoint/tensors/00321.safetensors
|
| 348 |
+
ef978f4506c8fb6d3b73805e4fd8ed2d0d48a22e3946141f7b00d5156bcf325a checkpoint/tensors/00322.safetensors
|
| 349 |
+
f7bd8b8db7bba066035323be0d08e70fd2e83ca8c76e2aed3b1a46ad513e403b checkpoint/tensors/00323.tensorbin
|
| 350 |
+
933b0bffbbdb807d74ecdaf5767dfa127bca1eff39fdf6796a5367a60bea2a21 checkpoint/tensors/00324.tensorbin
|
| 351 |
+
551ac18202925fc3768a79a9caab78cff2159be63fd8386c9b5a672401f65bf3 checkpoint/tensors/00325.tensorbin
|
| 352 |
+
70a7473f2e824b5f9f3b5bb91b102cd7483761643ca5c4d853a30c44901a84c6 checkpoint/tensors/00326.safetensors
|
| 353 |
+
ffe895a65a1e633964e3800f55084b76631d3c4e4443f42e234f912e8a93bfa3 checkpoint/tensors/00327.tensorbin
|
| 354 |
+
aa112ad32962cbc96ea14d2730f74c086ecc5511103f7608c3c7b442854bd9c9 checkpoint/tensors/00328.safetensors
|
| 355 |
+
fa5b50e415bc1e00096c802c8684b8667cec92d5dfbcfc663e9cf1b8b751f391 checkpoint/tensors/00329.safetensors
|
| 356 |
+
9e29e5361846363fca19daa4b33383c8b3d261a7e04e1da6c62960e14519ab61 checkpoint/tensors/00330.safetensors
|
| 357 |
+
e0f0f6fefd0da2b7df532eff6c74d4a0aeb0bea5e78cf53747d8ad47cb513425 checkpoint/tensors/00331.safetensors
|
| 358 |
+
b874bf3e0532ae654062a029f3b7c400ebf2801d87666531934f6e28a9f3b4e9 checkpoint/tensors/00332.safetensors
|
| 359 |
+
ce7270976c540756dc7068bd4d75aa9a93e4ae34ef41e8dd5636fc965a9ed86c checkpoint/tensors/00333.safetensors
|
| 360 |
+
6c79c48dcfeac132354cfbd904f6e2c7c8ecc459f293a98b5a5263a05d8a1f6a checkpoint/tensors/00334.safetensors
|
| 361 |
+
c40f618ec516c0395f72dd5dd11d042a079889ae7cb4571ac173e8df70cc0f7a checkpoint/tensors/00335.tensorbin
|
| 362 |
+
3bc65faaa82445b1751362a3cc1256c50b787c3245e7faec2886f664e0815bda checkpoint/tensors/00336.tensorbin
|
| 363 |
+
a4d560883f62ce9ce4158a2b7570ce0db3c4fec423bdb256cf7c898d3b1ed2d6 checkpoint/tensors/00337.tensorbin
|
| 364 |
+
f5526bcc2b454e1e2a2c3dc100deb7e1cba5ad544343188738e5742adbf2994f checkpoint/tensors/00338.safetensors
|
| 365 |
+
1f2f41055f41a8b19dae62716e2935d85b97639432ea0c5de79711884473772d checkpoint/tensors/00339.tensorbin
|
| 366 |
+
3e5c2370760a8dd78ff96c00be1af21786c71fe52abe1b5da870df375abe7a4c checkpoint/tensors/00340.safetensors
|
| 367 |
+
01d009782aef328fced6416467e1e081be23e908d7588f2942d43289e02976c2 checkpoint/tensors/00341.safetensors
|
| 368 |
+
e00cd3eeb3514e00e3f08a2ecfa4f2bb234313eec1d3fc79bf6ef94f0d7d89ab checkpoint/tensors/00342.safetensors
|
| 369 |
+
dba5657f19f3e88041f7fe81b47af97feb1fba9ca1db6afcc74e62000e962345 checkpoint/tensors/00343.safetensors
|
| 370 |
+
b8641836711dea2234c384222dee768e4a09ba15157fb2669c8e93ea5a8aec0e checkpoint/tensors/00344.safetensors
|
| 371 |
+
291d9e401610cd314415defe12396e50ac03946c3874514dddd61f3c0a2d4af6 checkpoint/tensors/00345.safetensors
|
| 372 |
+
f832ff73102e337143659675b3e4290294d533009104c8de059820c337f6890e checkpoint/tensors/00346.safetensors
|
| 373 |
+
3b1e2e5c1dbe7b6d7be94bf54cc112069b99e8d07c73b318c5f96b086c302a72 checkpoint/tensors/00347.tensorbin
|
| 374 |
+
d88e7b367a2709497cafa85edc350a74d8d5e61950920bef77d15d05ec9b849a checkpoint/tensors/00348.tensorbin
|
| 375 |
+
bf571a3f4add95848f757d01b1515e2fee5acbad831102d9eb034c24db025f6e checkpoint/tensors/00349.tensorbin
|
| 376 |
+
4104015aaa95de84c05cf0c9d47903b12d9730aebc61f3eb278476ef6b5bfd40 checkpoint/tensors/00350.safetensors
|
| 377 |
+
e83bb20e30a9aa8127b9f7c68cc1f6c028b734c314935aaf5806496fe25ee0ca checkpoint/tensors/00351.tensorbin
|
| 378 |
+
c982a6161cfd4b9fba51e79796c23d563bad2a546ef20b460b868e89d6bc5d11 checkpoint/tensors/00352.safetensors
|
| 379 |
+
8e1c3c64b4f87c2b0c4e778c6fb8c8a205174d18dc5ae521ae10484484da40d5 checkpoint/tensors/00353.safetensors
|
| 380 |
+
4043433e772192e6f4116c5342c11e12f312788b68b20dcb2e3bbc644ee37a19 checkpoint/tensors/00354.safetensors
|
| 381 |
+
947905523ce2e65dc2a20b27b8b808b93b28275642abb69f1add7fa6c5e9a3fe checkpoint/tensors/00355.safetensors
|
| 382 |
+
32957f1457668052016ae5f4a62d7599167d809fc8c4612b77a649602c7bc5ff checkpoint/tensors/00356.tensorbin
|
| 383 |
+
c1b9b6f9253a0e5d20e243ab5e677f77e01b702977f1a6b47482fd803c07538e checkpoint/tensors/00357.tensorbin
|
| 384 |
+
94dfcd51dc3aacbd2e2eebf82a26150c0da0fde7aa14aeab73728041e45c5a4a checkpoint/tensors/00358.safetensors
|
| 385 |
+
3ba7cb9bf833b8517324e58f01c59dee079ad852116ae318be5e5f5f1f4cb53c checkpoint/tensors/00359.tensorbin
|
| 386 |
+
6c617751b4eb1e98024d82cb229dcc42a00371c028b7725f21d6a1c1dc024fcf checkpoint/tensors/00360.safetensors
|
| 387 |
+
9072698b7bd9078c5660c71da724aea0ed39dde99385e67f31d9cf4f109b393b checkpoint/tensors/00361.safetensors
|
| 388 |
+
8eeb01b3c275dd175b45f409bc5e941f775bfb8786548ee16a9bd85301e2ccc7 checkpoint/tensors/00362.safetensors
|
| 389 |
+
d405ae9ab72b06cf9f8b652f5210d78b4fa84a06b46d953df06e3ccfdd877054 checkpoint/tensors/00363.safetensors
|
| 390 |
+
d5ece82c0ca9d91b1e39a1d283c6a1ce6e17244e02b713711d610ac2d5573cde checkpoint/tensors/00364.safetensors
|
| 391 |
+
d0c00f4f5ff5bebc87646776c35d265343c8c276db7ebdbb935531c8ca58af0e checkpoint/tensors/00365.safetensors
|
| 392 |
+
5a6ccc73efd577189cda08b2527afdeec98ae6a1a4776e346fe4b6596396cd34 checkpoint/tensors/00366.tensorbin
|
| 393 |
+
1bd2e60de97d3127d3c29e4331e6d0a58338e3b70055ff76d7c253b2b8df90f2 checkpoint/tensors/00367.tensorbin
|
| 394 |
+
74db066558ab1cfb1d09182226738a0c96af5a3bffe2494b6d4ac112e3b11918 checkpoint/tensors/00368.tensorbin
|
| 395 |
+
b64433e030f948f6a49268015c5d32df4ea8dbe86ae8d7b3b56ea054928d2317 checkpoint/tensors/00369.tensorbin
|
| 396 |
+
941cc537004d25fefb3be63583c7c1377f985af59d2fb9d347afaa8b85b9e63c checkpoint/tensors/00370.safetensors
|
| 397 |
+
02913ef3e17b7ceb8201daa2037f5d6566754265256d76949a3d22a5fb28c70f checkpoint/tensors/00371.tensorbin
|
| 398 |
+
51f18d486004b9f7619c117ea305098f805dfb92b686d3c28a9cf8898b157ec6 checkpoint/tensors/00372.safetensors
|
| 399 |
+
7e454367cf0d3f0d38057ca61b5bb5709993aae1fd6f435072e6afd2299eadbd checkpoint/tensors/00373.safetensors
|
| 400 |
+
eed68c136b8bb62ada3bdb30206d7cc39e520383de5f6886adcc6fb6fcc3129e checkpoint/tensors/00374.safetensors
|
| 401 |
+
88552be1e06d96166c05c9be44f058db4abbfcf4447f00bb0673de5154d384fd checkpoint/tensors/00375.safetensors
|
| 402 |
+
662afcbaa934fb721c55268fc622bdeb72962678d41c1174291f9682aa41bf98 checkpoint/tensors/00376.safetensors
|
| 403 |
+
5c5f8d0a1e900413a0a43736cb0e84c43b73eb341a5b8b9aaaac651fc0c6dd99 checkpoint/tensors/00377.safetensors
|
| 404 |
+
52bd7191382c0dc68dcb1ad3daf55729bbb6020fdb58f436765e707d1ef5197e checkpoint/tensors/00378.safetensors
|
| 405 |
+
e2b5fda61adc0cc32e36040203b19e91513cf0be439fba69bbda10519001f820 checkpoint/tensors/00379.tensorbin
|
| 406 |
+
500b05010b8e4dc06ae9af18df8ee2207c757bbb586cc34042400092086b75f4 checkpoint/tensors/00380.tensorbin
|
| 407 |
+
0ab06e612fb9dc83c0c81d8b31cd92ae183fba0d0e487f5607909e0dc1692fc1 checkpoint/tensors/00381.tensorbin
|
| 408 |
+
3fbc9604b5ee5c4f4747faa78f5247580735812c9258299eeae8ddf400f34ca3 checkpoint/tensors/00382.tensorbin
|
| 409 |
+
31dead960838bc8b8ebf535244fa3051f9cb4cf9b3a47a1570abcf92cf6edd4b checkpoint/tensors/00383.safetensors
|
| 410 |
+
3927c4cb0676bae29a88181561e508e5b22813866e293e130f8fac0a726c687a checkpoint/tensors/00384.tensorbin
|
| 411 |
+
118df08727745c490fae29eca4a550387a9cfa71d36d07f6d09b69a08291acb4 checkpoint/tensors/00385.safetensors
|
| 412 |
+
5b0e9d911642c1112a2514651b0402b3a18c9e0be6d3248ded286dbe3517f4f4 checkpoint/tensors/00386.safetensors
|
| 413 |
+
5f7c22a0e9b174d7aba458d5e583cec4e40ee00fd702d501fc6a7e01617ee371 checkpoint/tensors/00387.safetensors
|
| 414 |
+
300acb95b202f2c983a68083c9b73d07d6d102b22c8376a744bbb5b84487ee20 checkpoint/tensors/00388.safetensors
|
| 415 |
+
7a792110cd463a853c8d6cc7b7d4e62831243ca05e41ea409dd552902b8100bd checkpoint/tensors/00389.tensorbin
|
| 416 |
+
164680f101e3c2b1dba4136e806e90b798a9888ed6ba3101c6f1f4727bc5ef96 checkpoint/tensors/00390.tensorbin
|
| 417 |
+
af30c45b25e697ac32f6c988eda52e0a38d7de82b05705afdeb809a988eae5d6 checkpoint/tensors/00391.safetensors
|
| 418 |
+
f5ebb79485f41cfb35759d841f9a4dffa119b7b5718157474d871684e2ff6a9d checkpoint/tensors/00392.tensorbin
|
| 419 |
+
80b7e88c1170e38c3cd4c25e2b560d0e1d428699895ee53c8f5c13400447044f checkpoint/tensors/00393.safetensors
|
| 420 |
+
301a065ac6e6acb68b0601031d26992885b1476113b0685a595ae09339db329a checkpoint/tensors/00394.safetensors
|
| 421 |
+
cfa1f54448347bc7274da9155527a1fd5d2aea21467f92a52583fcc92792687d checkpoint/tensors/00395.safetensors
|
| 422 |
+
206f8d3a225c38e2b095632cb550471b8ea569c05295309f2c967761f9af3c4a checkpoint/tensors/00396.safetensors
|
| 423 |
+
c7197d3d5996e0041bc3d6e9d5eda77298ed76f9c53335e76e1880d14932386d checkpoint/tensors/00397.safetensors
|
| 424 |
+
06e82b3c15479f4c88020d5a9226d5ae5dd98437ab20e248abce79dab8d3f333 checkpoint/tensors/00398.safetensors
|
| 425 |
+
31c746a9713ac7adb43406d21bdaddd5263fe6b993a9fef7e06ac6c9c78c6601 checkpoint/tensors/00399.tensorbin
|
| 426 |
+
3906b06ce3787dc7207b533a5b3a1ed8ec18e272240b953c548ee07a3300b763 checkpoint/tensors/00400.tensorbin
|
| 427 |
+
6d2dcbd758f7498e083a4a0cfcfbacec0b9f3355ca484a364306e1dbebd8378b checkpoint/tensors/00401.tensorbin
|
| 428 |
+
a3dc095feecd7ec2fa99e23c3187ddfadced5769ec88e1850f2a5e26346c76e3 checkpoint/tensors/00402.safetensors
|
| 429 |
+
e3c8d5260b8b8b9bc3ef2a61cf2a082b3ea93c3c489956ce3b927397bf8aab5e checkpoint/tensors/00403.tensorbin
|
| 430 |
+
6ea56c96365c56669dee3859fc04a10733614d17c7ea7ca595a642c1aa9faef9 checkpoint/tensors/00404.safetensors
|
| 431 |
+
a20db5ccc60faa237e046ac4e7b97ed601caf04e6963c6e93ccad97787a4307a checkpoint/tensors/00405.safetensors
|
| 432 |
+
2221cccd72c09ae5f5772e700786133bea84ae41dbc602b217052923ffb3e862 checkpoint/tensors/00406.safetensors
|
| 433 |
+
a33b186f19e51b0d6916d426f9777a394648f9422cb1f8958460c0f6539c29b1 checkpoint/tensors/00407.safetensors
|
| 434 |
+
f080e2b5c63958ab128e21598ccf6e352aab1c2205d174de5bac92ca1d6cb49c checkpoint/tensors/00408.tensorbin
|
| 435 |
+
7945eead342178f8a6a21a8ac233a851e21915753c6459a093f0b70fcc3a2bb8 checkpoint/tensors/00409.tensorbin
|
| 436 |
+
387feace28cf17012298d482f7e8c68f29f1134760ac420b4ac3af4815ba0294 checkpoint/tensors/00410.safetensors
|
| 437 |
+
45ba03b35321b1296697f9949cb2d92a5ccb14f97cd022219955a68656664ae2 checkpoint/tensors/00411.tensorbin
|
| 438 |
+
640536bfed098ed2dc1efaae560a9b2c871e757d3affaa242e7ef468b6e48bd1 checkpoint/tensors/00412.safetensors
|
| 439 |
+
f52b1444dd4466d52873790066261d0c6d9d038b5ac269877e8aa029fecf9f86 checkpoint/tensors/00413.safetensors
|
| 440 |
+
bc59d7df0584793af27029b8e3854cefe2b83ef571332a3136679f402aebacbb checkpoint/tensors/00414.safetensors
|
| 441 |
+
1706db1da8443732217a5676dd60204f5d6d0f414f1914c6cc5aae37c2c4de70 checkpoint/tensors/00415.safetensors
|
| 442 |
+
392c7a06d0b106e9968a83b592e5fe87947dd1fecac749366c403a6ea2f87968 checkpoint/tensors/00416.safetensors
|
| 443 |
+
91837a639269ef3a0e6b79de043992c44a8d1c18f8135b32f8dabb953120be1a checkpoint/tensors/00417.safetensors
|
| 444 |
+
4871d3e8306bd7f58b283e6cb7928b89a938fc69063b870e65ba76cdb8424e89 checkpoint/tensors/00418.tensorbin
|
| 445 |
+
053dc7715cec657e0d5cb652264e708f36478143d014f9109e4dd00e619e6afa checkpoint/tensors/00419.tensorbin
|
| 446 |
+
d8c94968f8d4f8a2e8061547c2817980b01317dbbd36cc4ddcfb655c160899c1 checkpoint/tensors/00420.tensorbin
|
| 447 |
+
e96df26bb801b4f4e3d2c082aca40f71f170ecce2b8bea973ae89e8a94ffad23 checkpoint/tensors/00421.safetensors
|
| 448 |
+
e382563765bec4ec8cbacf7e76b4efcff8754346060d028a376ee92d4dd34fa2 checkpoint/tensors/00422.tensorbin
|
| 449 |
+
48134e5a383bbe269681f2f3db1bfa05eb170bb2b6147bf8a3069c7344e1d25f checkpoint/tensors/00423.safetensors
|
| 450 |
+
8d3fba88fdfac5577ad5af7c884037e18cea4cc144d6adcbf599631c415e5bae checkpoint/tensors/00424.safetensors
|
| 451 |
+
af1708fcc2b70e879e1d2a1a6a34b2ddb3c641520de4cd32f07c994632149b81 checkpoint/tensors/00425.safetensors
|
| 452 |
+
694e87094e8a7cfd313cf06efeceb7466a8e8329047657f2aa0a54edd8ffa70f checkpoint/tensors/00426.safetensors
|
| 453 |
+
e1ddcd72fd72b530f62ec5e3aa6fa3430ead9c244dbb0ece79999b013c10130b checkpoint/tensors/00427.safetensors
|
| 454 |
+
23edf639f94aed7e7be12f3f74429c2882a687d684cca369bf0db24ca30c54ec checkpoint/tensors/00428.safetensors
|
| 455 |
+
778afebc18173a5dce8402a0b5604bb6f6742d1955646c0f7bd14e3dd898580d checkpoint/tensors/00429.safetensors
|
| 456 |
+
2a1e41d055c7de741e24cda79161514f2da045f19cec7a4d48a768f951aa912c checkpoint/tensors/00430.tensorbin
|
| 457 |
+
774112aad52efa04b4fdc3ddf0ed076be65dde26f04b86100a78f81741157b6b checkpoint/tensors/00431.tensorbin
|
| 458 |
+
47df8a1a5675944c9303f42625019aae5f0841c99ddf17860894d594aab36038 checkpoint/tensors/00432.tensorbin
|
| 459 |
+
3e80e9dbbf2727bc184a1b4eb15f71356e7ef1c2f1229783f702f0d6aed33552 checkpoint/tensors/00433.safetensors
|
| 460 |
+
135ea11b8de2629a6f40f522c84788059cfa98130b830061430fea3aec7c787e checkpoint/tensors/00434.tensorbin
|
| 461 |
+
0a6337eaad23da3aa923c7e8d5daa77074cc56310816fad1d5006d01adcac494 checkpoint/tensors/00435.safetensors
|
| 462 |
+
700be95730fdcd5e74c14c95c0c94ea5ecf2623ddccc85a970984e84a624b3fa checkpoint/tensors/00436.safetensors
|
| 463 |
+
9bba0fb6d5f8af7c13b7d67b49c5cbd3905f6d3e7c26d9c1cd66d6cf68882ec3 checkpoint/tensors/00437.safetensors
|
| 464 |
+
b0a65b896c7b373d8e7a03988b6672257a388b5c6772e03cd00b1b9f03114192 checkpoint/tensors/00438.safetensors
|
| 465 |
+
b9a8efaff04fb39d75fef7764ba8f49912dea994d8a8108f0aaef65a0bd18794 checkpoint/tensors/00439.safetensors
|
| 466 |
+
1ae5291a5aabf3c0f0f9550eca59e8e31f3ff207dd6970013e64d7942c4a763f checkpoint/tensors/00440.safetensors
|
| 467 |
+
399c86adfde976e0e25f3c9d505f73b084009442691ee4367481be58b4fad362 checkpoint/tensors/00441.safetensors
|
| 468 |
+
40e734a3a15cfa5507375abec138af46d0afb9faad64eb3af5f8e26a352c3f17 checkpoint/tensors/00442.tensorbin
|
| 469 |
+
5e78bfe351466486c03dbdc39bf313cae53a02f73de524f031b52c1ea01eeac4 checkpoint/tensors/00443.tensorbin
|
| 470 |
+
04b1e6a84915a11b329af92c6c613197175a1f7525bc557dfb25354b613f2801 checkpoint/tensors/00444.tensorbin
|
| 471 |
+
11dcdcf629a2742b8c6bfb4b9e4c43bfc13d38d8d36731936e134b54dbb909d0 checkpoint/tensors/00445.safetensors
|
| 472 |
+
bbbca23d75a833abd42bb4098543fd4f9551e79ac95b5f4c62484649d1045d40 checkpoint/tensors/00446.tensorbin
|
| 473 |
+
c90fd44c1107b7ea201921e34843a97892c84f139711385267004a9aefe98d75 checkpoint/tensors/00447.safetensors
|
| 474 |
+
8cbfde46a4d81445531ac1b5d63658d64bfab0fc2ef082f4268293fde80c8ff4 checkpoint/tensors/00448.safetensors
|
| 475 |
+
1cc0f34276ab4c78a805746665b6e6a521d48b59c6b6bf36f148c670357061c2 checkpoint/tensors/00449.safetensors
|
| 476 |
+
6e1bdc5a53f77d45a2290cb8ea8d70d01f25e9fa83caf231be8cf75076be9bb6 checkpoint/tensors/00450.safetensors
|
| 477 |
+
1c95d49c2be1e13bc3a24678964c65a5d272d55ddafe4553f91adeb46f77427a checkpoint/tensors/00451.tensorbin
|
| 478 |
+
1cbb3896127a3198474fa676b5ebbce1c29661c21ad4bf1cba48321e2cc54926 checkpoint/tensors/00452.tensorbin
|
| 479 |
+
285cfbf38bc47939ffbcb160ecda125244c1b6965b13c40613ad022c924ed6a9 checkpoint/tensors/00453.safetensors
|
| 480 |
+
593ec9941c04b5edbf977cad929c94c1d9fc1b9d1344fda12b58aa9094f07c52 checkpoint/tensors/00454.tensorbin
|
| 481 |
+
da0079b7208305a16e15d313663664d5ed2791222c564aa86b5997d3159b3400 checkpoint/tensors/00455.safetensors
|
| 482 |
+
c15aa81c67f0b93b9499736cc2673a758e1d70d6fee5e40ad36aeb08470b8943 checkpoint/tensors/00456.safetensors
|
| 483 |
+
a5539158c3981a309745499ae97672289e74c8e65c699d6d73738f83130e8ced checkpoint/tensors/00457.safetensors
|
| 484 |
+
d0c4abd04267fecebb649ec29e44be04c7ee91e0d3204ea7109d860bf0db1024 checkpoint/tensors/00458.safetensors
|
| 485 |
+
9193c28379462d1ac32fe1b20a72fe4a454335feb9294be3aba0d85188cdaa0d checkpoint/tensors/00459.safetensors
|
| 486 |
+
da9b133a750c06487cbca457f05dbd82eeb8e5ecf197131a93d327a96a818061 checkpoint/tensors/00460.safetensors
|
| 487 |
+
d31609a8b7a38f97bb6644ce7cc03770bff60aa70eff2da09ab1e8fd9badedcb checkpoint/tensors/00461.tensorbin
|
| 488 |
+
bab52f6b8b71225acbe9c4d745f6596bdbbddaad3ed703f7997352d0474486f5 checkpoint/tensors/00462.tensorbin
|
| 489 |
+
2cb6c902e41c5be476832d9968a54f782413d66886f895f4e5c90e370e5ec019 checkpoint/tensors/00463.tensorbin
|
| 490 |
+
0824112bdd5eb270f2bd0301ebcee7d5340b01cc1d059434f919fd471c44f81e checkpoint/tensors/00464.safetensors
|
| 491 |
+
ecbb4066b48a19e253585003f2663b8c927e8134e1d10a1d5a36cbd7ff6cfb33 checkpoint/tensors/00465.tensorbin
|
| 492 |
+
8d4b04e7bce60b341f1766b0d033d5d81f688f93332183027d9ab41db2c46edd checkpoint/tensors/00466.safetensors
|
| 493 |
+
00b95e504637624cd68a051fe2d59f7137a8468f954eea19737c54db84a6094d checkpoint/tensors/00467.safetensors
|
| 494 |
+
193a468fe0ecad48521114ca47bdadef25d2cce812941be3ebf56bb298e2b60f checkpoint/tensors/00468.safetensors
|
| 495 |
+
fcc62f682a7cb9347018ddd5282b42811d2654b22d24a3a62583b4f36ff03f52 checkpoint/tensors/00469.safetensors
|
| 496 |
+
fe3930c3ccb32d37743f85334f0f57c11da99ceea98067b7a5ab92de2e4b86ac checkpoint/tensors/00470.safetensors
|
| 497 |
+
cfae1c10334a0b4a3be13ae8d64af96385254ef5880357a603625e20c90c0086 checkpoint/tensors/00471.safetensors
|
| 498 |
+
1d168adb24ca87be74b2a87f3305e6fa99525bd46890ba79638ef1f21ca7673e checkpoint/tensors/00472.safetensors
|
| 499 |
+
5910c57587e07814685b50563d1cb8f35c70fa8e2b81aab35ce60f0bfa88a257 checkpoint/tensors/00473.tensorbin
|
| 500 |
+
2bc158467526bb8a20dbaea1747882f0500b4c469ab8a59ba70f4280c594e6a7 checkpoint/tensors/00474.tensorbin
|
| 501 |
+
d5d708f47a07eed8432f60be221f9f399bd9e001bf6d7521d49bdb115e2781bc checkpoint/tensors/00475.tensorbin
|
| 502 |
+
56e12160bb4dc11b44c67b6221dcf9cb6d0324e27265a2ea8c94e526c701dac2 checkpoint/tensors/00476.safetensors
|
| 503 |
+
de597c0d4097bc9d781ecb9495d6e2243e3b358761241a3e22e08bd3706b66f1 checkpoint/tensors/00477.tensorbin
|
| 504 |
+
36b228da14b05f9ffffedd56a2c0dc803bd1f7d3fb15ba1e88e7b7f6f4bed281 checkpoint/tensors/00478.safetensors
|
| 505 |
+
e5d76fe208825dfca5c7b250ce42bf509b9952548d20b7dd99887e605a510945 checkpoint/tensors/00479.safetensors
|
| 506 |
+
a6e326dc337f0091cdc99f4648f5908263ed1b9f15f2fdcd5e4c08682165421f checkpoint/tensors/00480.safetensors
|
| 507 |
+
59f79dd68f363379678e50b084042f0fddd7758c7d71f8f544ec370d31a1eb31 checkpoint/tensors/00481.safetensors
|
| 508 |
+
62fa7d289f0fff7a4a048c886e47dcd0b1c33a22406da68b279489527e4d7d37 checkpoint/tensors/00482.safetensors
|
| 509 |
+
9a5b54e0147aab4b620af2e2ea7bbc5d9224af8aa03247eb3e7ff05624427a1c checkpoint/tensors/00483.safetensors
|
| 510 |
+
ff347414f813f32d570f527ab249278f2219770fe5ee7257e3ef8b0475cb8422 checkpoint/tensors/00484.safetensors
|
| 511 |
+
6a0855fe02d8fc491b2122a86a348948a70992049afec25a55097aa6618221eb checkpoint/tensors/00485.safetensors
|
| 512 |
+
7c3710f7605adb64f1d6b0d759f241c44e5b5696b2bf72aa7375ff933e059f9e checkpoint/tensors/00486.safetensors
|
| 513 |
+
d5df4e469de20f888596de1e9fbc91a43315af3a9277d2008e4805df15eba8cb checkpoint/tensors/00487.safetensors
|
| 514 |
+
551934f9c06a2afdc68792a22f7e558de59ca9517ba90fb448f7df7c2553502f checkpoint/tensors/00488.safetensors
|
| 515 |
+
4529c477928e9f6ed72fffe2f573daec108b18245d4f4a39639458c675aaff56 checkpoint/tensors/00489.safetensors
|
| 516 |
+
5f9e4d4901a92b997e463c1f46055088b6cca5ca61a6522d1b9f64c4bb81cb42 checkpoint/tokenizer.json
|
| 517 |
+
316230d6a809701f4db5ea8f8fc862bc3a6f3229c937c174e674ff3ca0a64ac8 checkpoint/tokenizer_config.json
|
| 518 |
+
7768af27c1fafa9cc9011c1dc20067e03f8915e03b63504550e11d5066986d13 checkpoint/video_preprocessor_config.json
|
| 519 |
+
ce99b4cb2983d118806ce0a8b777a35b093e2000a503ebde25853284c9dfa003 checkpoint/vocab.json
|
| 520 |
+
8e7b3e30bf5e74ed59d90b0b2e4e0a33a4985c7f7504fd7bd75d97a96087c5dc collect_importance.py
|
| 521 |
+
cfe6517ca2f1d45e1bced935ec796ed1a1c44a46affe231403be92a8286e783e evaluation/assistant-masks.json
|
| 522 |
+
1d99d13c9e19bed67ded85dd7ff475873a0094546fffb75d18b358fa1a740523 evaluation/behavior.jsonl
|
| 523 |
+
7afb1723a477daebd26e8c9b8391d6a081bf4d8a0ba1930b8a840131faaaf144 evaluation/evaluation-protocol.json
|
| 524 |
+
a53a6e18c4d33b787f15bb1aad6ce88de97f25239ad2236617a62256b297f979 evaluation/instruction-eval.jsonl
|
| 525 |
+
5b14998390eb10d78ebcb3869f6710ef72f41ff708889228165c335d816f144b evaluation/instruction-provenance.json
|
| 526 |
+
3e0ef42f2092b094473afc38e7f9cdab1eea1a4cd6d77895d3367f16259723e4 evaluation/long-context.jsonl
|
| 527 |
+
62bb803e55d143e602a660993fbf9c363de1069aa1776ec4335dbd06353587dd evaluation/selection-policy.json
|
| 528 |
+
edf56b67f604561433d5125ebbcfbb85be8a9e2d34cd14a8f94d0e459ae120de evaluation/trajectories.jsonl
|
| 529 |
+
66772c7c01029d27a8d07012ecb76aad94da936f22fc76d472b5c5f08764ebc9 evidence/mtp/arc-challenge-shard1.json
|
| 530 |
+
bab23aa359297da7dde575591e9c0d432c893ae3b6fbc5e47cf35884327a1836 evidence/mtp/benchmark.py
|
| 531 |
+
d152b6f3df69200c3dfad4f821025ab668e01452ae4c74fb978b514f4cf32820 evidence/mtp/equivalence-summary.json
|
| 532 |
+
3cac0a57ab13127fc28e0591cfe312c11487a127fa17e91a0dc8a8420aedeb1d evidence/mtp/fixed-runtime-provenance.json
|
| 533 |
+
362b10ebd943ecd0414f34ef8de02c7dadca34c9b59d19de288cab7497da9950 evidence/mtp/gsm8k-shard1.json
|
| 534 |
+
87fa288b6276af0056f27a6fa426950d4f2ede848190b33d1abafdb6953d1517 evidence/mtp/lmx-native-mtp-off.json
|
| 535 |
+
355038cae1eff485d1be6c503ce8cfc545a58f531f4bc43b6542ef8f263a45ca evidence/mtp/lmx-native-mtp-on.json
|
| 536 |
+
844d8b92bef08dd734da0d26c61b6b5d50d64ec7e5aa3c7a04cda0a52b445ae3 evidence/mtp/native-mtp-off-speed.json
|
| 537 |
+
d672e598b15f47248aff35c754663f8970807e4b0c019f77e1e456fd462d8973 evidence/mtp/native-mtp-on-speed.json
|
| 538 |
+
156aa4f50f1e786520d97e1a0b9ba4e379402df5a42f05e06bb0c4778c501de2 evidence/mtp/native-off-serving-command.json
|
| 539 |
+
05018f6d3279b86db3a32fdf0ef5eb9d7b35d88f76c2b2cf0072afcd2682d9d0 evidence/mtp/native-on-serving-command.json
|
| 540 |
+
fc3983f9954c25afaccf5cfb47df7dd1992446901bb4925ccae86719732c5561 evidence/mtp/quality-commands.json
|
| 541 |
+
5f580b6eedad4a81f954cf6924ff75ce3de73db460f399d4a383864795b3e11a evidence/mtp/quality-protocol.json
|
| 542 |
+
86dc22ff70cdf20a4ffdfc24c05a0a3ab18c8c354f1dc005f6e8857024a789dd evidence/mtp/quality-request-traces.jsonl
|
| 543 |
+
ab9345abf7492617cd1310dc87b306e47b854887454ae3f63fa60bbd4a90526c evidence/mtp/request-isolation-fixed.json
|
| 544 |
+
539e89c3f67ff9cb219bdc34dbde0ccf98718786593305b0bcde0a5ce185be04 evidence/mtp/verify_equivalence.py
|
| 545 |
+
6416311f11f5aee55f077f1818f61460078468d9b27cdab7fefb304aa4dfc037 evidence/previous-no-mtp-quality/native-heldout-all-token-quality.json
|
| 546 |
+
98e6018710a5fb7496bb725178b25f08490f5a6eeae6ecba3f65990921508094 evidence/previous-no-mtp-quality/native-heldout-assistant-quality.json
|
| 547 |
+
5fbf430d311f9556adaf3952397e161fae0517f7737c13916011fa2fd5da6f08 evidence/previous-no-mtp-quality/precision-summary.json
|
| 548 |
+
83661fe8f501a9620b9c63c1c1b5824f7fd82b7802665db1f5c91ce45de5508a evidence/previous-no-mtp-quality/quality-summary.json
|
| 549 |
+
25e6c97b125e3fc969f9ea99290e1d3ca43db93720b699eb46f21224d8dd3c9d native_checkpoint.py
|
| 550 |
+
31f51ad1d7bc8cac7032d8a827856b8e31b5682b008b6daa41a5e3b9625548b6 precision-plan.json
|
| 551 |
+
802887a19d7e543272f061928124ca9691c6c2e745576c198ad3bbba5ccda94a precision-summary.json
|
| 552 |
+
860b81602f93bb01da7a095fc1f976d6b94c8ace623c4759848152a8ccdd7f8b prepare_assistant_masks.py
|
| 553 |
+
881ed010e90c3658497b1efa3bd844f6ab4647d1d5af848b6838416832acdd53 prepare_behavior.py
|
| 554 |
+
2862bfd2a04d894374e290698b9f888d9bf8e74f2e80b2053a7db660bea144d8 prepare_corpus.py
|
| 555 |
+
c5abd94571e74b281cff61c342c5037742e5384ed3a478ce6c9c3df7d42fc5fa prepare_instruction_corpus.py
|
| 556 |
+
865a3003363aaa5aa23aec452acd4db717ca347884f1ed8fc9e6e115b4e4ae27 quality-summary.json
|
| 557 |
+
4082a9f8e76f3eff57579a13eea20cbd862ad1c200dac6f0ba15105bc2be875b rank_precision.py
|
| 558 |
+
286b970571921b3a4ca3aa80a75a55c4191de404b4c95ad84828349bc504e046 reference_eval.py
|
| 559 |
+
b64d3d1fdebd07f6c7f60930829c80327f16d469a4470a5ec523b1c1ae1d2ce3 reference_generate.py
|
| 560 |
+
0d7ea076699733ed08d20519dc75d62e342b4dd43d53c9dacef2d1d08dd6b50b release-manifest.json
|
| 561 |
+
c558d575ac98af21215d86fcbc01f6d7c3ffd7c69907964635cf59424f08c0de reproduction.json
|
| 562 |
+
804cd90bad7ecc9e23b17b014897b957f5114d153839d528b1ac8a4b265689a3 runtime/Dockerfile
|
| 563 |
+
d03d6d273119df3a6a0794e49ec8e880b7945a219308a9227d2454a9da78139a runtime/Dockerfile.incremental
|
| 564 |
+
a6cba85bc92e0cff7a450b1d873c0eaa2e9fc96bf472df0247a26bec77bf3ff9 runtime/licenses/tt-metal/LICENSE
|
| 565 |
+
a729a2fb26ff46f60cf69d9903889236c22672f843041c651ae24263d753abb3 runtime/licenses/tt-metal/LICENSE_understanding.txt
|
| 566 |
+
387a33040fa411a419e684d93df255822193dbfab5cdbdc88fe6acc9bb1f096d runtime/licenses/tt-metal/NOTICE
|
| 567 |
+
c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4 runtime/licenses/vllm/LICENSE
|
| 568 |
+
94cd328bc9bda593791da19ee2b51f442f24352937136a93d139116aa6251c7a runtime/overlay/home/container_app_user/tt-metal/models/common/rmsnorm.py
|
| 569 |
+
32c19dcdf83d6d1315911c9418c86d125d68690032ba307dc7d2d8f166930046 runtime/overlay/home/container_app_user/tt-metal/models/experimental/gated_attention_gated_deltanet/tt/ttnn_delta_rule_ops.py
|
| 570 |
+
05e839857238de477ef6f8c19f2429b50a77b58b8c9982b894013fc8d541faf7 runtime/overlay/home/container_app_user/tt-metal/models/experimental/gated_attention_gated_deltanet/tt/ttnn_delta_rule_seq.py
|
| 571 |
+
4c191baf202017cb99545647b92ad2722dd16166d0709e874021ee09c9d02af8 runtime/overlay/home/container_app_user/tt-metal/models/experimental/gated_attention_gated_deltanet/tt/ttnn_gated_attention.py
|
| 572 |
+
ae5ee8836003861bcba2ccd2b01da369c3f875a4fed33a7a26d5566a86fee1a8 runtime/overlay/home/container_app_user/tt-metal/models/experimental/gated_attention_gated_deltanet/tt/ttnn_gated_deltanet.py
|
| 573 |
+
a4c45d3f54f37735ebe412255b92efa76479abf7083f88cfb71d27e404b001ad runtime/overlay/home/container_app_user/tt-metal/ttnn/cpp/ttnn/operations/matmul/device/factory/matmul_multicore_reuse_mcast_1d_program_factory.cpp
|
| 574 |
+
35367672e01f5d494dbd3d7dc62d8fcce020ae870788b1e3f4ecfbaff7e1b358 runtime/overlay/home/container_app_user/tt-metal/ttnn/cpp/ttnn/operations/matmul/device/kernels/compute/bmm_large_block_zm_fused_bias_activation.cpp
|
| 575 |
+
9d2c601e9f3c630d8039c922f218ad012de8049e79785d7de31f6c95889a384a runtime/overlay/home/container_app_user/tt-metal/ttnn/cpp/ttnn/operations/matmul/device/kernels/dataflow/reader_bmm_tile_layout_in1_sender_writer_padding.cpp
|
| 576 |
+
f413d42d8ede06cad0a8f811ebd139a65623726735cf1d8bc9f9dc8b68585d3a runtime/overlay/home/container_app_user/tt-metal/ttnn/cpp/ttnn/operations/matmul/device/kernels/dataflow/reader_bmm_tile_layout_qwen_lmhead_packed.cpp
|
| 577 |
+
479cc7812419ab87a7b8b1d838aeb635417abb7707522f5fce90b2d5c331f298 runtime/overlay/home/container_app_user/tt-metal/ttnn/cpp/ttnn/operations/transformer/chunk_gated_delta_rule/chunk_gated_delta_rule.cpp
|
| 578 |
+
14c171d5ad013dbddac971acdc63858cae66394b3af2e0c96a79111f0210605b runtime/overlay/home/container_app_user/tt-metal/ttnn/cpp/ttnn/operations/transformer/chunk_gated_delta_rule/chunk_gated_delta_rule.hpp
|
| 579 |
+
fa9d781a98bc0691c731f481d9b034e698cfb4bbd3bf4155e1cffbdc30361329 runtime/overlay/home/container_app_user/tt-metal/ttnn/cpp/ttnn/operations/transformer/chunk_gated_delta_rule/chunk_gated_delta_rule_nanobind.cpp
|
| 580 |
+
2afeccc3d26969736c39cfe965e8ee8075b7bffc556b25c5ba665ccaeff9b711 runtime/overlay/home/container_app_user/tt-metal/ttnn/cpp/ttnn/operations/transformer/chunk_gated_delta_rule/device/chunk_gated_delta_rule_device_operation.cpp
|
| 581 |
+
c48587954dbf056c24f36753cfa1b2d1d1d556a9fd2b4bdebc78eaa04f43ecfe runtime/overlay/home/container_app_user/tt-metal/ttnn/cpp/ttnn/operations/transformer/chunk_gated_delta_rule/device/chunk_gated_delta_rule_device_operation.hpp
|
| 582 |
+
abd6b46952266668ae26fc030af9b2e940f23c38d92e6742af07e636284ea48b runtime/overlay/home/container_app_user/tt-metal/ttnn/cpp/ttnn/operations/transformer/chunk_gated_delta_rule/device/chunk_gated_delta_rule_device_operation_types.hpp
|
| 583 |
+
9e5062cfe6ecd903ae89b70109decd0102387b8d81cf70f4f3e7303849a529bd runtime/overlay/home/container_app_user/tt-metal/ttnn/cpp/ttnn/operations/transformer/chunk_gated_delta_rule/device/chunk_gated_delta_rule_program_factory.cpp
|
| 584 |
+
6c6cf9bccada84cc245666291adbbcf899f7ed4e58a7d05e361af8169c79aca1 runtime/overlay/home/container_app_user/tt-metal/ttnn/cpp/ttnn/operations/transformer/chunk_gated_delta_rule/device/chunk_gated_delta_rule_program_factory.hpp
|
| 585 |
+
1a5814a4dec3a35020bee4a73d2b191b561e55a4c16c197cce81972205de251c runtime/overlay/home/container_app_user/tt-metal/ttnn/cpp/ttnn/operations/transformer/chunk_gated_delta_rule/device/kernels/compute/recurrent_gated_delta_rule.cpp
|
| 586 |
+
e638d13cb4c63a9f11622495145efe95b7254936fedcdaac89caa538672252a4 runtime/overlay/home/container_app_user/tt-metal/ttnn/cpp/ttnn/operations/transformer/chunk_gated_delta_rule/device/kernels/dataflow/reader_recurrent_gated_delta_rule.cpp
|
| 587 |
+
434c56acbc87aa486c8a9d7630b9eadf01b5f1ef32f8c279999c4000ee90e84f runtime/overlay/home/container_app_user/tt-metal/ttnn/cpp/ttnn/operations/transformer/chunk_gated_delta_rule/device/kernels/dataflow/writer_recurrent_gated_delta_rule.cpp
|
| 588 |
+
41599bcd4bd9d9d58137031d6e20723c2389443b43963d2597864ec36917251f runtime/overlay/home/container_app_user/tt-metal/ttnn/cpp/ttnn/operations/transformer/gated_delta_attn/device/gated_delta_attn_program_factory.cpp
|
| 589 |
+
36045c0bceb322faeb32ff3d4d30d4af1cc1bf2e5d6ae0555e413db34ea3c241 runtime/overlay/home/container_app_user/tt-metal/ttnn/cpp/ttnn/operations/transformer/gated_delta_attn/device/kernels/compute/gated_delta_attn.cpp
|
| 590 |
+
2b1b54a9a70d9c82ef5e58e0e636d5c27c4bb84f9d8fa071c65676c666c8e2cc runtime/overlay/home/container_app_user/vllm/plugins/vllm-tt-plugin/src/vllm_tt_plugin/model_input.py
|
| 591 |
+
8eec17ddf95709c77b0d569c5e64c0a040c448c03fd0910206b77b3a0026e297 runtime/overlay/home/container_app_user/vllm/plugins/vllm-tt-plugin/src/vllm_tt_plugin/model_runner.py
|
| 592 |
+
77a555aedc9ff4f77b85d65a540b20d1195c2c61d12215b84eef56b7053fff37 runtime/overlay/home/container_app_user/vllm/plugins/vllm-tt-plugin/src/vllm_tt_plugin/platform.py
|
| 593 |
+
5c8c71bd386a34c13b992f6c57e89c10f6eac59843581750da9959b72420fb8d runtime/overlay/home/container_app_user/vllm/plugins/vllm-tt-plugin/src/vllm_tt_plugin/worker.py
|
| 594 |
+
e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855 runtime/qwen36/__init__.py
|
| 595 |
+
e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855 runtime/qwen36/demo/__init__.py
|
| 596 |
+
98e1c2924c7387748bccb4db7ab4e4b7186747a281145cc7dca29576e44d8f6c runtime/qwen36/demo/benchmark_vision.py
|
| 597 |
+
6387caf56cef80bb8f1ac6288c58a40954fb73715cf608ae60b364bc1c496026 runtime/qwen36/demo/sample_prompts/eval_frankenstein_long.json
|
| 598 |
+
f244039a6f6d0cfc82d0b166c4af0635b31420acfdef5e68e6063b94ab7fdc8e runtime/qwen36/demo/sample_prompts/input_data_long_4k.json
|
| 599 |
+
7e001aeee4fd25da5ba1586d285d190ffda01d32292758b5748bb4ea2afeab32 runtime/qwen36/demo/sample_prompts/vision_demo.json
|
| 600 |
+
861ba5cb5489cf31f8e6aa3563d0177f56cb4348b07d10663def680d95c41787 runtime/qwen36/demo/sample_prompts/vision_multi_image.json
|
| 601 |
+
94c2a3f11d8b199de287efd895b5842ea4432adb616e42173888943cfbbdabdf runtime/qwen36/demo/sample_prompts/vision_text_only.json
|
| 602 |
+
e30102476ad9bede85e95cddbdaf56e5ab6d210ff531089887f059abe89e49a6 runtime/qwen36/demo/sample_prompts/vision_video.json
|
| 603 |
+
9e3c1d86391312efb3ef6c8f07636d586322e7d22359cfb8d58a9cb1443fc37d runtime/qwen36/demo/text_demo.py
|
| 604 |
+
fe25dd9e81c3fc8b2ef3908b613500fd4de84106d9e2005ab753f5b62821f523 runtime/qwen36/demo/vision_demo.py
|
| 605 |
+
e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855 runtime/qwen36/tests/__init__.py
|
| 606 |
+
facc72bc2d4daaa6063efb448360b48d48e936388df0bd9ad84ec1833b85f6b2 runtime/qwen36/tests/conftest.py
|
| 607 |
+
3c67a83e7c64acdcf7341ca3ac75a2fff7c2c288daa1ae16a4b969cb57d26b29 runtime/qwen36/tests/pcc_thresholds.json
|
| 608 |
+
faac4db41e15dea443bb9185635e94d5c9e968fbb8e4b95009ce874f88192566 runtime/qwen36/tests/test_attention_tp.py
|
| 609 |
+
b4b122dfeeb530d460a7bdb20534621009710d3b52bbfb28d475581107303422 runtime/qwen36/tests/test_decode_bucketing.py
|
| 610 |
+
ddbbfd6ab73db4bfc9b45ae32bc6de44d5d032e8d41cb8d27f39aa0a8ab01d73 runtime/qwen36/tests/test_factory.py
|
| 611 |
+
4fc8a93b4c122f24ffa7c15be459011b39ff72e0ffe5a3665453507f2a74b8d8 runtime/qwen36/tests/test_gdn_tp.py
|
| 612 |
+
2cc65458dd3acc06de9a8da5a2457e9d1dae639dabf7436b456f632fdf6408bd runtime/qwen36/tests/test_generate_tp.py
|
| 613 |
+
0c7a32c00de7d42700e78e63a12b5abe8d11ecc41f36a33d8b078ffd7f788d2e runtime/qwen36/tests/test_mlp.py
|
| 614 |
+
3402b0165e6dedd386d9b57a23df80811c6f4f261c818f4a5e448b04608a245a runtime/qwen36/tests/test_mlp_tp.py
|
| 615 |
+
123d57bfe23fd7d2d2f070b62589b1af617733dfb566695b11ee775523c29080 runtime/qwen36/tests/test_model.py
|
| 616 |
+
d3733c41e470165895dce249ecbf86776427bb5f9348f267988bcfce7a1a568b runtime/qwen36/tests/test_model_tp.py
|
| 617 |
+
893db166c3aea884e05a2099529e4842e03f2ae3ce45fdeae9b765b502c053a4 runtime/qwen36/tests/test_patch_merger.py
|
| 618 |
+
b4d0ac8c72656e8dc771bf2a9c231f8dded674a9e5ff14be82e878b3d85f6072 runtime/qwen36/tests/test_prefill.py
|
| 619 |
+
cf0c1fa054bd79f639cfe773b0acf4f6bcf53bf01c77911a886064e85fdd6a47 runtime/qwen36/tests/test_rope_tp.py
|
| 620 |
+
7444ccb142551cdf4190305039b94d7d4622a050dd905a53da3256e8c985c2b5 runtime/qwen36/tests/test_sampling.py
|
| 621 |
+
e04c1814d95959aee63d7a6fdc2032762418f1f1956a31bd0dbfaeff033f27dd runtime/qwen36/tests/test_vision_attention.py
|
| 622 |
+
3065caa4c032cc89140986418d1c9ae4cfe06ad8e91b05de950579b17dcb3e6c runtime/qwen36/tests/test_vision_block.py
|
| 623 |
+
beb49e1a62ab665523329fed2da45481803676acd98b705c7c2f5be420f28710 runtime/qwen36/tests/test_weight_mapping.py
|
| 624 |
+
eddaca444bc1bea1c73aa28299739a910192b305b7bdc686fb09f96b1767ac2a runtime/qwen36/tests/test_wrapped_model.py
|
| 625 |
+
e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855 runtime/qwen36/tests/unit/__init__.py
|
| 626 |
+
7dcbd7c08346a18f41f7f22c8fd63503468bcf69b1c4b6d60737cb02081c770d runtime/qwen36/tests/unit/conftest.py
|
| 627 |
+
bc4ff5c0c429bca4607534329e62cbcd84061d0b1a9a02dc7ee6534db47cb1e7 runtime/qwen36/tests/unit/test_attention.py
|
| 628 |
+
66607195f94c141b153f198e0310dd8a3c20a50e3b7689e9d31b19b1e18e3558 runtime/qwen36/tests/unit/test_embedding.py
|
| 629 |
+
15323b3051d4c466d1a6021681ea37b0e70806358ec619f69419e2b36426b501 runtime/qwen36/tests/unit/test_gdn.py
|
| 630 |
+
d6babb72837bec309e047d4c03e91733a1711dad2c0568fc65fb15c6c1d94385 runtime/qwen36/tests/unit/test_layer.py
|
| 631 |
+
327f70e8d720f4ca52327ecbc87ec52e8d5969b24ab1bd593916362c51604994 runtime/qwen36/tests/unit/test_lm_head.py
|
| 632 |
+
3947b52bd589f1daa46634f70f0e0f47c2715d7d75c05607e8c6222e049ba49f runtime/qwen36/tests/unit/test_mlp.py
|
| 633 |
+
5a17473efbce8b70ddcb323dfe10a36208a1ec5e069c72298f9d5a0423343ba6 runtime/qwen36/tests/unit/test_model.py
|
| 634 |
+
1196bed05216527986a366bd98985ccf0ec04bb6bd692dd5c30b7c6f4017f38e runtime/qwen36/tests/unit/test_rms_norm.py
|
| 635 |
+
c22883d945579f7365ef50fbfd942c62329e2a4f6a907187d8a56151b57eef0c runtime/qwen36/tests/unit/test_rope.py
|
| 636 |
+
101215ebc50ea587700c4f6dd64ad6fbd40101efc81a917991de856fc67e607e runtime/qwen36/tests/unit/test_substate.py
|
| 637 |
+
e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855 runtime/qwen36/tt/__init__.py
|
| 638 |
+
86fbe666fb79e0b726789175fca14912c6bfb971295355165e97a3e8eed4a991 runtime/qwen36/tt/attention/__init__.py
|
| 639 |
+
6bdec51939dd9824cdb0c904cae46de82fb148d6f8f4be7ecf6d727f5c0cc28e runtime/qwen36/tt/attention/config.py
|
| 640 |
+
73ccc4511896933cc0b40bac9b43c0eec583e747a10b4af622e6916233a115d5 runtime/qwen36/tt/attention/decode.py
|
| 641 |
+
42b9480eb138c3c5628d55432e19d875d6fc1847be9ca46b271979d2194017eb runtime/qwen36/tt/attention/gated_attention.py
|
| 642 |
+
1cc028e66492bf13d6d8e8040026c6ae4c82618bab19f1db8a7681929553ad77 runtime/qwen36/tt/attention/prefill.py
|
| 643 |
+
b841d519864e702ef0c9b63306a28df8d57af6ad17462b9e282b6912239f6e70 runtime/qwen36/tt/attention/rope_tp.py
|
| 644 |
+
a31ec742f2382af56bb620812859be7c25e675f1115c12e9ab713bc2d8bf0f3e runtime/qwen36/tt/attention/tp.py
|
| 645 |
+
bd50f4f9bcd820fb44a82293e5c34842c9bad55701600410137d19a0e8033428 runtime/qwen36/tt/attention/weights.py
|
| 646 |
+
4fb0b62e54f4da4df2013bb05a4504cbf35563e6c7d7ef1073740dff328e4289 runtime/qwen36/tt/common.py
|
| 647 |
+
0c0d8c94369c6e4b5d246afb1f8d36ccaff4b9510b093a76bd243c1c0b67bf51 runtime/qwen36/tt/fused_swiglu.py
|
| 648 |
+
e4d8f6233ec1e86fb80fb0032e01e9c5f7c48ade6603fca50ad59db1c6b5b7f9 runtime/qwen36/tt/fused_swiglu_kernel.cpp
|
| 649 |
+
d27e043b36083293a4841c5f69109aeb89bdbe01e1c11cc00e118af04e8bb24a runtime/qwen36/tt/gateup_layout/gateup_layout_reader.cpp
|
| 650 |
+
cd9dade74044020df60b3158c82c7ced484a8012d150ae87dcadd09fe45c82e8 runtime/qwen36/tt/gdn/__init__.py
|
| 651 |
+
d36dc3dabace49f020e96d3979ffd7fba415e9d244db7a511e2ea6611b14e306 runtime/qwen36/tt/gdn/config.py
|
| 652 |
+
56dd072ad942f2f59af543521cb3a3981bdddd906976892a9c2c9e821f366644 runtime/qwen36/tt/gdn/decode.py
|
| 653 |
+
d3721fd705f5dcf513ad90fe21496fe2e475ffb14683a07533f51aba77762d7a runtime/qwen36/tt/gdn/fused_chunk.py
|
| 654 |
+
2a248f57b59a74e5a4ef04ab1e6386f24de4ddfdbff2de3904f6abbc1d2168c0 runtime/qwen36/tt/gdn/fused_commit.py
|
| 655 |
+
f92493aaefe3ca27f8e0006b0f749ce900ccd63f4a37a261a126cebda619b1b1 runtime/qwen36/tt/gdn/fused_commit_kernel.cpp
|
| 656 |
+
10f0f96fc1a4f9a8888ff606974f4f5ad0309f119be30fb5490c79e1a4584e5c runtime/qwen36/tt/gdn/fused_verify_output.cpp
|
| 657 |
+
e782e7711a02eac5de100bcd2b26b463437d0f6ad05ee3b345e319c4c1dff3ce runtime/qwen36/tt/gdn/fused_verify_output.py
|
| 658 |
+
f38160d095b1fb384a875f2dbd949d5cd0708f8316d708a9e10d3798b41bfad6 runtime/qwen36/tt/gdn/fused_verify_writer.cpp
|
| 659 |
+
5b3b1efad65cf9ddf6c201fff6ced22dda0f8feb924d564047edbb484ca0f006 runtime/qwen36/tt/gdn/gated_deltanet.py
|
| 660 |
+
45531e65d90e79ddf8d51b7757060ff8e86aaf1b3e80e0f2740c90f7c363a11a runtime/qwen36/tt/gdn/native_frontend.py
|
| 661 |
+
0d46fe1db394499da78dda8c6f83f9776692388b55afa2ff3cf615d23f4490b3 runtime/qwen36/tt/gdn/native_frontend_kernel.cpp
|
| 662 |
+
4961e26b7fb0ecd068ec92871389d5502e244f0002dda8fdbec68da5deaab508 runtime/qwen36/tt/gdn/native_output.py
|
| 663 |
+
16088ec9cd0d2001081f58cf595668e5e565f1db0a3f6a93ec654eba965335bd runtime/qwen36/tt/gdn/native_output_kernel.cpp
|
| 664 |
+
34a2323786d33640ad9fb8a9d74f8e8f830a8a46e3946c5029648c5647ed944e runtime/qwen36/tt/gdn/state.py
|
| 665 |
+
72c1a2fe5d7b60b6740d42037c70ff2fe93a5e1fbaa3b5e18d629e2f4cd47de9 runtime/qwen36/tt/gdn/tp.py
|
| 666 |
+
a7dbd253ab080dac1885e1a6f1bd7831032c2222c7200932c179941f61112302 runtime/qwen36/tt/gdn/weights.py
|
| 667 |
+
5eb6f5c1874e88ae097f6be4a884b38790525fa42f62ca88c3763ef490b3c456 runtime/qwen36/tt/generator_interface.py
|
| 668 |
+
fa986f11ee9ca7b22e4387b5b47ac936192513ac96e0c633df266671fddae4fd runtime/qwen36/tt/layer.py
|
| 669 |
+
9a9356f88f6124b74967e1fed34bea5383a83ffbf7206e944a0964e70ba1a11f runtime/qwen36/tt/mlp.py
|
| 670 |
+
23d23caef5aac2f34b3ac41f9af9b5cd65ae0a67b7424e3d50d9741fc919e684 runtime/qwen36/tt/model.py
|
| 671 |
+
268444cc9edcd069e0fa74141a286035512551c40141e50565726d134cc7903c runtime/qwen36/tt/model_config.py
|
| 672 |
+
6ccd78a5ffd176e7267d60d6cff960014918df203130614a0cd2b201eeffa4c8 runtime/qwen36/tt/mtp.py
|
| 673 |
+
2e3bc11507f4fbba00464593e048faa1e29d8928722143839541c96e319be7fc runtime/qwen36/tt/qwen36_vllm.py
|
| 674 |
+
5bb650d835e5ed6cdfa375e64c80ed92a1758996883ffb7c238b2195869f6f00 runtime/qwen36/tt/rms_norm.py
|
| 675 |
+
baa056d37129f1fa97a444ba5f6d5f3a5cfd6c15402dcfad13d300b11f2cef26 runtime/qwen36/tt/rope.py
|
| 676 |
+
bb43f0cde336c3f84725d47a64ed2b506b5287bdd0e910cd24b13feed0a0826a runtime/qwen36/tt/tp_common.py
|
| 677 |
+
cc2c73418c0211e5db739ce19e47b7c479f507d246e37239d90d0f469d73f1ef runtime/qwen36/tt/vision/__init__.py
|
| 678 |
+
02a48050552879762da15634627500c04e49c620ef880c05c794715c536e5511 runtime/qwen36/tt/vision/functional.py
|
| 679 |
+
92f33dcd670d8bf2000a8f523b2dcba4cada9599d309e0d7d079a87720742f91 runtime/qwen36/tt/vision/model.py
|
| 680 |
+
c7af2a141d45c8d7843445aaa0f68be9eaf5f6249d1c42c2a8f0fa393496aae9 runtime/qwen36/tt/vision/patch_merger.py
|
| 681 |
+
5f99bb9243dce7c37045f82d5917b005bb7aa6ca6dfac788bb69abc44927a0be runtime/qwen36/tt/vision/vision_attention.py
|
| 682 |
+
1d951f0732445ee099112711c8a59b9701a235e35668b87b97cb965faa74ad64 runtime/qwen36/tt/vision/vision_block.py
|
| 683 |
+
47fd0fcf240b4a1dba399e2f608b050ecc789ffe2c7871d375b94574bda49a23 runtime/qwen36/tt/vision/vision_distributed_layernorm.py
|
| 684 |
+
cb2b91f9d6cdf788847871eeb3fd21059dc3934ed274e61e40f5e9fc98e0db60 runtime/qwen36/tt/vision/vision_layernorm.py
|
| 685 |
+
6c3c092c33df7cba0483b6ab88903f29a95f2fcecc723d3884e3cbe4effcd0e5 runtime/qwen36/tt/vision/vision_mlp.py
|
| 686 |
+
bf1eb6f5d47243c46c33ee2c3506f6985131a379161d71a02b7bff88bdec5a4d runtime/qwen36/tt/vision/vision_model_config.py
|
| 687 |
+
d7ad88592fb7de07d87fa0d10f9d1bd0d4dcc92856438fffc394f90e29e650d2 runtime/qwen36/tt/weight_mapping.py
|
| 688 |
+
e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855 runtime/qwen36/utils/__init__.py
|
| 689 |
+
813aa26dc9053e1217425a3cacf0c63255254f9d17723f3b5ac378167cfca28c runtime/qwen36/utils/substate.py
|
| 690 |
+
2a1f7f46c7a4096cc327921ad2e3b7fb1d0d5859a531a5c6d4bc56aa0978050b runtime/source-manifest.json
|
| 691 |
+
a87126812d5e9c761a7670589175e3be6b855f60df76a641bf2691789f7899a1 score_behavior.py
|
| 692 |
+
680552951c60918e48e6f652f59fc5ed75473c663f1eaa63e116979f6a6a43d9 score_logits.py
|
| 693 |
+
59d35218b4780c71e90f2a7cf148cc1f8a0db44a4a5c5dbd90cda624d6b54b1a selected-overrides.json
|
| 694 |
+
0118c611f52574043f9fe3b9a32657e59a663b5e92e051f3159e91fa0e1b1e01 serve_native.py
|
| 695 |
+
2939042f81d63dd6d90b1b13574c2a505623a07c1684bb9e3d048182c25eed59 tt_eval.py
|
api_validate.py
ADDED
|
@@ -0,0 +1,125 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"""Exercise native serving behavior, near-limit prefill and greedy decode over HTTP.
|
| 2 |
+
Local diagnostics, not an official task benchmark or Unsloth Divergence-300.
|
| 3 |
+
"""
|
| 4 |
+
import argparse
|
| 5 |
+
import json
|
| 6 |
+
import statistics
|
| 7 |
+
import time
|
| 8 |
+
import urllib.request
|
| 9 |
+
from pathlib import Path
|
| 10 |
+
from score_behavior import passed
|
| 11 |
+
|
| 12 |
+
|
| 13 |
+
def payload(model, messages, max_tokens):
|
| 14 |
+
return {"model": model, "messages": messages, "temperature": 0, "top_p": 1,
|
| 15 |
+
"seed": 9472, "max_tokens": max_tokens,
|
| 16 |
+
"chat_template_kwargs": {"enable_thinking": False}}
|
| 17 |
+
|
| 18 |
+
|
| 19 |
+
def send(base_url, body):
|
| 20 |
+
return urllib.request.urlopen(urllib.request.Request(
|
| 21 |
+
base_url.rstrip('/') + '/v1/chat/completions',
|
| 22 |
+
data=json.dumps(body).encode(), headers={"Content-Type": "application/json"}), timeout=600)
|
| 23 |
+
|
| 24 |
+
|
| 25 |
+
def check_case(base_url, model, case, max_tokens):
|
| 26 |
+
messages = case.get('messages', [{"role": "user", "content": case['prompt']}])
|
| 27 |
+
body = payload(model, messages, max_tokens)
|
| 28 |
+
if case.get('tools'):
|
| 29 |
+
body.update(tools=case['tools'], tool_choice='auto')
|
| 30 |
+
started = time.monotonic()
|
| 31 |
+
with send(base_url, body) as response:
|
| 32 |
+
result = json.load(response)
|
| 33 |
+
message = result['choices'][0]['message']
|
| 34 |
+
text = message.get('content') or ''
|
| 35 |
+
if case['check'] == 'weather_tool':
|
| 36 |
+
calls = message.get('tool_calls') or []
|
| 37 |
+
ok = len(calls) == 1 and calls[0]['function']['name'] == 'get_weather'
|
| 38 |
+
if ok:
|
| 39 |
+
arguments = json.loads(calls[0]['function']['arguments'])
|
| 40 |
+
ok = arguments.get('city', '').casefold() == str(case['expected']).casefold()
|
| 41 |
+
else:
|
| 42 |
+
ok = passed(case, text)
|
| 43 |
+
return {'id': case['id'], 'pass': bool(ok), 'seconds': time.monotonic() - started,
|
| 44 |
+
'response': result}
|
| 45 |
+
|
| 46 |
+
|
| 47 |
+
def benchmark(base_url, model):
|
| 48 |
+
body = payload(model, [{"role": "user", "content":
|
| 49 |
+
"Write a detailed technical explanation of why batch-one language-model decoding "
|
| 50 |
+
"is often memory-bandwidth limited. Explain weight traffic, activation traffic, "
|
| 51 |
+
"kernel launch overhead, and how these differ from prompt prefill. Use several paragraphs."}], 128)
|
| 52 |
+
body.update(stream=True, stream_options={'include_usage': True})
|
| 53 |
+
started = time.monotonic()
|
| 54 |
+
first_token, finished, usage = None, None, None
|
| 55 |
+
pieces = []
|
| 56 |
+
with send(base_url, body) as response:
|
| 57 |
+
for raw in response:
|
| 58 |
+
line = raw.decode().strip()
|
| 59 |
+
if not line.startswith('data: '):
|
| 60 |
+
continue
|
| 61 |
+
data = line[6:]
|
| 62 |
+
if data == '[DONE]':
|
| 63 |
+
break
|
| 64 |
+
event = json.loads(data)
|
| 65 |
+
if event.get('usage'):
|
| 66 |
+
usage = event['usage']
|
| 67 |
+
for choice in event.get('choices', []):
|
| 68 |
+
delta = choice.get('delta', {})
|
| 69 |
+
text = (delta.get('reasoning_content') or '') + (delta.get('content') or '')
|
| 70 |
+
if text:
|
| 71 |
+
if first_token is None:
|
| 72 |
+
first_token = time.monotonic()
|
| 73 |
+
pieces.append(text)
|
| 74 |
+
if choice.get('finish_reason') is not None:
|
| 75 |
+
finished = time.monotonic()
|
| 76 |
+
if first_token is None or finished is None or not usage or usage['completion_tokens'] < 2:
|
| 77 |
+
raise ValueError('Missing streamed timing or endpoint token-usage evidence')
|
| 78 |
+
return {'text': ''.join(pieces), 'usage': usage,
|
| 79 |
+
'ttft_ms': (first_token - started) * 1000,
|
| 80 |
+
'decode_tokens_per_second': (usage['completion_tokens'] - 1) / (finished - first_token),
|
| 81 |
+
'decode_seconds': finished - first_token,
|
| 82 |
+
'timing_scope': 'Client-observed first nonempty delta through finish event; excludes first output token'}
|
| 83 |
+
|
| 84 |
+
|
| 85 |
+
def main():
|
| 86 |
+
parser = argparse.ArgumentParser(description=__doc__)
|
| 87 |
+
parser.add_argument('--base-url', default='http://127.0.0.1:8001')
|
| 88 |
+
parser.add_argument('--model', default='Qwen/Qwen3.5-9B')
|
| 89 |
+
parser.add_argument('--cases', type=Path, required=True)
|
| 90 |
+
parser.add_argument('--long-case', type=Path, required=True)
|
| 91 |
+
parser.add_argument('--output', type=Path, required=True)
|
| 92 |
+
args = parser.parse_args()
|
| 93 |
+
report = {'status': 'running', 'scope': __doc__, 'base_url': args.base_url,
|
| 94 |
+
'model': args.model, 'behavior': [], 'long_context': [], 'decode_runs': []}
|
| 95 |
+
try:
|
| 96 |
+
report['initial_decode'] = benchmark(args.base_url, args.model)
|
| 97 |
+
for case in map(json.loads, args.cases.read_text().splitlines()):
|
| 98 |
+
result = check_case(args.base_url, args.model, case, 64)
|
| 99 |
+
report['behavior'].append(result)
|
| 100 |
+
print(json.dumps({'case': result['id'], 'pass': result['pass']}), flush=True)
|
| 101 |
+
for case in map(json.loads, args.long_case.read_text().splitlines()):
|
| 102 |
+
result = check_case(args.base_url, args.model, case, 16)
|
| 103 |
+
result['prepared_prompt_tokens'] = len(case['token_ids'])
|
| 104 |
+
report['long_context'].append(result)
|
| 105 |
+
print(json.dumps({'case': result['id'], 'pass': result['pass']}), flush=True)
|
| 106 |
+
for _ in range(3):
|
| 107 |
+
report['decode_runs'].append(benchmark(args.base_url, args.model))
|
| 108 |
+
report.update(status='complete', behavior_passes=sum(r['pass'] for r in report['behavior']),
|
| 109 |
+
long_context_passes=sum(r['pass'] for r in report['long_context']),
|
| 110 |
+
median_decode_tokens_per_second=statistics.median(r['decode_tokens_per_second'] for r in report['decode_runs']),
|
| 111 |
+
identical_greedy_completions=len({r['text'] for r in report['decode_runs']}) == 1)
|
| 112 |
+
report['request_isolation_passed'] = all(
|
| 113 |
+
run['text'] == report['initial_decode']['text'] for run in report['decode_runs'])
|
| 114 |
+
if not report['request_isolation_passed']:
|
| 115 |
+
raise RuntimeError('Greedy output changed after interleaved requests, including long-context prefill')
|
| 116 |
+
except BaseException as error:
|
| 117 |
+
report.update(status='failed', error=str(error))
|
| 118 |
+
raise
|
| 119 |
+
finally:
|
| 120 |
+
args.output.write_text(json.dumps(report, indent=2, ensure_ascii=False) + '\n')
|
| 121 |
+
print(json.dumps({k: v for k, v in report.items() if k not in ('behavior', 'long_context', 'decode_runs')}, indent=2), flush=True)
|
| 122 |
+
|
| 123 |
+
|
| 124 |
+
if __name__ == '__main__':
|
| 125 |
+
main()
|
build_native_checkpoint.py
ADDED
|
@@ -0,0 +1,277 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
#!/usr/bin/env python3
|
| 2 |
+
"""Stream original Qwen3.5-9B text and MTP safetensors into a standalone TT-native PTQ.
|
| 3 |
+
|
| 4 |
+
Run only with exclusive TT device ownership. TTNN host conversion
|
| 5 |
+
initializes device metadata. Example inside the pinned container:
|
| 6 |
+
python /work/build_native_checkpoint.py --weights /weights/Qwen3.5-9B \
|
| 7 |
+
--output /work/native-candidate --profile current-bfp4 \
|
| 8 |
+
--overrides /work/candidate.json --importance /work/importance/importance.npz \
|
| 9 |
+
--container-image IMAGE_ID --device-ownership-confirmed
|
| 10 |
+
|
| 11 |
+
Each target tensor is remapped individually; linear matrices are transposed
|
| 12 |
+
BEFORE native TILE quantization. Other tensors, including BF16 embeddings and
|
| 13 |
+
original FP32 nonlinear parameters, are preserved as individual safetensors.
|
| 14 |
+
All original mtp.* tensors bypass remapping and quantization, preserving their
|
| 15 |
+
runtime keys, values and source dtype for the existing one-layer MTP runtime.
|
| 16 |
+
Importance only measures/ranks precision error: it does not optimize rounding,
|
| 17 |
+
implement an imatrix-aware quantizer, or measure output KL. Candidate quality
|
| 18 |
+
must be evaluated on held-out text with the separate TT runner/scorer.
|
| 19 |
+
"""
|
| 20 |
+
import argparse
|
| 21 |
+
import datetime
|
| 22 |
+
import importlib.util
|
| 23 |
+
import json
|
| 24 |
+
import platform
|
| 25 |
+
import shutil
|
| 26 |
+
from pathlib import Path
|
| 27 |
+
|
| 28 |
+
from native_checkpoint import DTYPES, FORMAT, MANIFEST, MTP_KEYS, digest, local_file, matrix_family, save_json, tensor_hash, tensor_precision, validate_mtp_keys
|
| 29 |
+
|
| 30 |
+
ROOT = Path(__file__).resolve().parent
|
| 31 |
+
ASSETS = ("config.json", "tokenizer.json", "tokenizer_config.json", "vocab.json", "merges.txt",
|
| 32 |
+
"chat_template.jinja", "special_tokens_map.json", "added_tokens.json", "generation_config.json",
|
| 33 |
+
"tokenizer.model", "preprocessor_config.json", "processor_config.json",
|
| 34 |
+
"video_preprocessor_config.json")
|
| 35 |
+
|
| 36 |
+
|
| 37 |
+
def load_remapper(path):
|
| 38 |
+
spec = importlib.util.spec_from_file_location("native_export_weight_mapping", path)
|
| 39 |
+
module = importlib.util.module_from_spec(spec)
|
| 40 |
+
spec.loader.exec_module(module)
|
| 41 |
+
return module.remap_qwen36_state_dict
|
| 42 |
+
|
| 43 |
+
|
| 44 |
+
def weight_error(original, restored, importance, module, torch):
|
| 45 |
+
"""Bounded row chunks; diagonal input second moments, not output KL."""
|
| 46 |
+
moments, count = None, None
|
| 47 |
+
if importance is not None and module + ".sumsq" in importance.files:
|
| 48 |
+
import numpy as np
|
| 49 |
+
|
| 50 |
+
sumsq = importance[module + ".sumsq"]
|
| 51 |
+
count = int(importance[module + ".count"].item())
|
| 52 |
+
if count < 1 or sumsq.shape != (original.shape[1],) or not np.isfinite(sumsq).all() or (sumsq < 0).any():
|
| 53 |
+
raise ValueError(f"Invalid activation importance for {module}")
|
| 54 |
+
moments = torch.from_numpy(sumsq.copy()).to(torch.float64) / count
|
| 55 |
+
error_sum, original_sum, weighted_error, weighted_signal, max_error = 0.0, 0.0, 0.0, 0.0, 0.0
|
| 56 |
+
for start in range(0, original.shape[0], 128):
|
| 57 |
+
source = original[start:start + 128].to(torch.float32)
|
| 58 |
+
error = restored[start:start + 128].to(torch.float32) - source
|
| 59 |
+
squared = error.square()
|
| 60 |
+
error_sum += squared.sum(dtype=torch.float64).item()
|
| 61 |
+
original_sum += source.square().sum(dtype=torch.float64).item()
|
| 62 |
+
max_error = max(max_error, error.abs().max().item())
|
| 63 |
+
if moments is not None:
|
| 64 |
+
weighted_error += (squared.sum(dim=0, dtype=torch.float64) * moments).sum().item()
|
| 65 |
+
weighted_signal += (source.square().sum(dim=0, dtype=torch.float64) * moments).sum().item()
|
| 66 |
+
result = {"sum_squared_error": error_sum, "mean_squared_error": error_sum / original.numel(),
|
| 67 |
+
"max_abs_error": max_error,
|
| 68 |
+
"relative_squared_error": error_sum / original_sum if original_sum else None,
|
| 69 |
+
"importance_available": moments is not None}
|
| 70 |
+
if moments is not None:
|
| 71 |
+
result.update({"activation_rows": count, "importance_weighted_squared_error": weighted_error,
|
| 72 |
+
"importance_weighted_mean_per_output": weighted_error / original.shape[0],
|
| 73 |
+
"importance_weighted_relative_squared_error": weighted_error / weighted_signal if weighted_signal else None})
|
| 74 |
+
return result
|
| 75 |
+
|
| 76 |
+
|
| 77 |
+
def export(args):
|
| 78 |
+
import numpy as np
|
| 79 |
+
import torch
|
| 80 |
+
import ttnn
|
| 81 |
+
from safetensors import safe_open
|
| 82 |
+
from safetensors.torch import save_file
|
| 83 |
+
from tt_eval import precision_map
|
| 84 |
+
|
| 85 |
+
torch.set_num_threads(args.cpu_threads)
|
| 86 |
+
weights, output = args.weights.resolve(), args.output.resolve()
|
| 87 |
+
if not output.is_relative_to(ROOT) or output == ROOT or output.is_relative_to(weights) or weights.is_relative_to(output):
|
| 88 |
+
raise ValueError(f"Output must be isolated beneath {ROOT}, outside original weights")
|
| 89 |
+
if output.exists() and any(output.iterdir()):
|
| 90 |
+
raise ValueError("Refusing to overwrite a nonempty output directory")
|
| 91 |
+
config = json.loads((weights / "config.json").read_text())
|
| 92 |
+
text = config["text_config"]
|
| 93 |
+
if (text["num_hidden_layers"], text["hidden_size"], text["vocab_size"]) != (32, 4096, 248320):
|
| 94 |
+
raise ValueError("Only original Qwen3.5-9B is supported")
|
| 95 |
+
overrides = json.loads(args.overrides.read_text()) if args.overrides else None
|
| 96 |
+
precision = precision_map(args.profile, overrides, text["layer_types"])
|
| 97 |
+
remap_path = ROOT / "runtime/qwen36/tt/weight_mapping.py"
|
| 98 |
+
remap = load_remapper(remap_path)
|
| 99 |
+
index_path = weights / "model.safetensors.index.json"
|
| 100 |
+
index = json.loads(index_path.read_text())["weight_map"]
|
| 101 |
+
validate_mtp_keys(index)
|
| 102 |
+
shards = {}
|
| 103 |
+
for name, filename in index.items():
|
| 104 |
+
if name.startswith("mtp.") or name == "lm_head.weight" or (name.startswith("model.language_model.") and ".mtp." not in name and ".visual." not in name):
|
| 105 |
+
shards.setdefault(filename, []).append(name)
|
| 106 |
+
if not shards:
|
| 107 |
+
raise ValueError("Expected original model.language_model text checkpoint keys")
|
| 108 |
+
importance = np.load(args.importance, allow_pickle=False) if args.importance else None
|
| 109 |
+
output.mkdir(parents=True, exist_ok=True)
|
| 110 |
+
(output / "tensors").mkdir()
|
| 111 |
+
(output / "provenance").mkdir()
|
| 112 |
+
files, tensors, source_shards, errors = {}, {}, {}, {}
|
| 113 |
+
|
| 114 |
+
def register(path):
|
| 115 |
+
path.chmod(0o644)
|
| 116 |
+
name = str(path.relative_to(output))
|
| 117 |
+
files[name] = {"bytes": path.stat().st_size, "sha256": digest(path)}
|
| 118 |
+
return name
|
| 119 |
+
|
| 120 |
+
for name in ASSETS:
|
| 121 |
+
source = weights / name
|
| 122 |
+
if source.is_file():
|
| 123 |
+
shutil.copyfile(source, output / name)
|
| 124 |
+
register(output / name)
|
| 125 |
+
licenses = [p for p in weights.iterdir() if p.is_file() and p.name.upper().startswith(("LICENSE", "NOTICE", "COPYING"))]
|
| 126 |
+
if not licenses or not all((output / name).is_file() for name in ("config.json", "tokenizer.json", "tokenizer_config.json")):
|
| 127 |
+
raise ValueError("Original config/tokenizer/license assets are required for standalone export")
|
| 128 |
+
for source in licenses:
|
| 129 |
+
shutil.copyfile(source, output / source.name)
|
| 130 |
+
register(output / source.name)
|
| 131 |
+
for source, destination in ((Path(__file__), output / "provenance/build_native_checkpoint.py"),
|
| 132 |
+
(ROOT / "native_checkpoint.py", output / "native_checkpoint.py"),
|
| 133 |
+
(remap_path, output / "provenance/weight_mapping.py"),
|
| 134 |
+
(ROOT / "tt_eval.py", output / "provenance/tt_eval.py"),
|
| 135 |
+
(ROOT / "precision-plan.json", output / "provenance/precision-plan.json"),
|
| 136 |
+
(index_path, output / "provenance/original.safetensors.index.json")):
|
| 137 |
+
shutil.copyfile(source, destination)
|
| 138 |
+
register(destination)
|
| 139 |
+
if args.overrides:
|
| 140 |
+
shutil.copyfile(args.overrides, output / "provenance/overrides.json")
|
| 141 |
+
register(output / "provenance/overrides.json")
|
| 142 |
+
if args.importance:
|
| 143 |
+
shutil.copyfile(args.importance, output / "provenance/importance.npz")
|
| 144 |
+
register(output / "provenance/importance.npz")
|
| 145 |
+
if (args.importance.parent / "metadata.json").is_file():
|
| 146 |
+
shutil.copyfile(args.importance.parent / "metadata.json", output / "provenance/importance-metadata.json")
|
| 147 |
+
register(output / "provenance/importance-metadata.json")
|
| 148 |
+
counter, quantized, fp32, mtp_fp32 = 0, 0, 0, 0
|
| 149 |
+
for filename, names in sorted(shards.items()):
|
| 150 |
+
shard = local_file(weights, filename)
|
| 151 |
+
source_shards[filename] = {"bytes": shard.stat().st_size, "sha256": digest(shard)}
|
| 152 |
+
with safe_open(str(shard), framework="pt", device="cpu") as source:
|
| 153 |
+
for original_name in sorted(names):
|
| 154 |
+
original = source.get_tensor(original_name)
|
| 155 |
+
original_hash = tensor_hash(original)
|
| 156 |
+
is_mtp = original_name.startswith("mtp.")
|
| 157 |
+
remapped = {original_name: original} if is_mtp else remap({original_name: original})
|
| 158 |
+
for name, value in remapped.items():
|
| 159 |
+
if name in tensors or "visual" in name or ("mtp" in name and not is_mtp):
|
| 160 |
+
raise ValueError(f"Unexpected duplicate/non-text remapped tensor: {name}")
|
| 161 |
+
dtype_name = None if is_mtp else tensor_precision(name, precision)
|
| 162 |
+
entry = {"source_name": original_name, "source_shard": filename,
|
| 163 |
+
"source_shape": list(original.shape), "source_torch_dtype": str(original.dtype),
|
| 164 |
+
"source_tensor_sha256": original_hash, "shape": list(value.shape),
|
| 165 |
+
"family": matrix_family(name), "restored_torch_dtype": str(value.dtype)}
|
| 166 |
+
if dtype_name is None:
|
| 167 |
+
value_hash = original_hash if is_mtp else tensor_hash(value)
|
| 168 |
+
path = output / "tensors" / f"{counter:05d}.safetensors"
|
| 169 |
+
save_file({name: value.contiguous()}, str(path))
|
| 170 |
+
with safe_open(str(path), framework="pt", device="cpu") as saved:
|
| 171 |
+
restored = saved.get_tensor(name)
|
| 172 |
+
if (restored.dtype != value.dtype or not torch.equal(restored, value)
|
| 173 |
+
or (is_mtp and tensor_hash(restored) != original_hash)):
|
| 174 |
+
raise ValueError(f"Lossless roundtrip failed: {name}")
|
| 175 |
+
del restored
|
| 176 |
+
if is_mtp:
|
| 177 |
+
mtp_fp32 += int(value.dtype == torch.float32)
|
| 178 |
+
else:
|
| 179 |
+
fp32 += int(value.dtype == torch.float32)
|
| 180 |
+
entry.update({"storage": "safetensors-lossless", "tensor_sha256": value_hash,
|
| 181 |
+
"roundtrip": {"exact_values": True, "exact_dtype": True}})
|
| 182 |
+
else:
|
| 183 |
+
if value.ndim != 2 or value.dtype != torch.bfloat16:
|
| 184 |
+
raise ValueError(f"Expected original BF16 linear matrix: {name} {value.shape} {value.dtype}")
|
| 185 |
+
path = output / "tensors" / f"{counter:05d}.tensorbin"
|
| 186 |
+
oriented = value.T.contiguous()
|
| 187 |
+
native = ttnn.from_torch(oriented, dtype=getattr(ttnn, DTYPES[dtype_name]), layout=ttnn.TILE_LAYOUT)
|
| 188 |
+
del oriented
|
| 189 |
+
ttnn.dump_tensor(str(path), native)
|
| 190 |
+
reloaded = ttnn.load_tensor(str(path))
|
| 191 |
+
rounded = ttnn.to_torch(reloaded).to(torch.bfloat16)
|
| 192 |
+
if not torch.equal(ttnn.to_torch(native), rounded):
|
| 193 |
+
raise ValueError(f"Native serialization or BF16 host restoration changes values: {name}")
|
| 194 |
+
del native, reloaded
|
| 195 |
+
restored = rounded.T.contiguous()
|
| 196 |
+
# Deliberately exercise the same transpose/copy that runtime converters do.
|
| 197 |
+
second = ttnn.from_torch(restored.T.contiguous(), dtype=getattr(ttnn, DTYPES[dtype_name]), layout=ttnn.TILE_LAYOUT)
|
| 198 |
+
second_path = output / "tensors" / f"{counter:05d}.roundtrip.tensorbin"
|
| 199 |
+
ttnn.dump_tensor(str(second_path), second)
|
| 200 |
+
exact_values = torch.equal(rounded, ttnn.to_torch(second))
|
| 201 |
+
exact_bytes = digest(path) == digest(second_path)
|
| 202 |
+
if not exact_values or not exact_bytes:
|
| 203 |
+
raise ValueError(f"Native requantization is not idempotent: {name}; values={exact_values}, bytes={exact_bytes}")
|
| 204 |
+
second_path.unlink()
|
| 205 |
+
del second, rounded
|
| 206 |
+
errors[name] = {"source_module": original_name.removesuffix(".weight"), "precision": dtype_name,
|
| 207 |
+
**weight_error(value, restored, importance, original_name.removesuffix(".weight"), torch)}
|
| 208 |
+
del restored
|
| 209 |
+
entry.update({"storage": "ttnn-tile", "precision": dtype_name,
|
| 210 |
+
"native_dtype": DTYPES[dtype_name], "native_shape": [value.shape[1], value.shape[0]],
|
| 211 |
+
"orientation": "input,output", "layout": "TILE_LAYOUT",
|
| 212 |
+
"roundtrip": {"exact_values": exact_values, "exact_serialized_bytes": exact_bytes}})
|
| 213 |
+
quantized += 1
|
| 214 |
+
entry["file"] = register(path)
|
| 215 |
+
entry["sha256"] = files[entry["file"]]["sha256"]
|
| 216 |
+
tensors[name] = entry
|
| 217 |
+
counter += 1
|
| 218 |
+
print(json.dumps({"tensor": name, "precision": dtype_name or "lossless", "bytes": files[entry["file"]]["bytes"]}), flush=True)
|
| 219 |
+
del original, remapped, value
|
| 220 |
+
if importance is not None:
|
| 221 |
+
importance.close()
|
| 222 |
+
if not {"tok_embeddings.weight", "output.weight", "norm.weight"} <= tensors.keys() or fp32 != 48:
|
| 223 |
+
raise ValueError(f"Missing top-level tensors or original FP32 nonlinear tensors: fp32={fp32}, expected 48")
|
| 224 |
+
validate_mtp_keys(tensors)
|
| 225 |
+
save_json(output / "quantization-error.json", {"method": "Unmodified TTNN rounding; importance-weighted precision ranking only, not optimized values or output KL",
|
| 226 |
+
"formula": "sum_out,in ((W-Wq)^2 * input_sumsq[in]/input_count)",
|
| 227 |
+
"tensors": errors})
|
| 228 |
+
register(output / "quantization-error.json")
|
| 229 |
+
manifest = {"format": FORMAT, "schema_version": 1, "scope": "text-only-no-vision-with-mtp",
|
| 230 |
+
"mtp": {"enabled": True, "num_speculative_tokens": 1,
|
| 231 |
+
"source_tensor_count": len(MTP_KEYS), "storage": "lossless"},
|
| 232 |
+
"status": "tensor_roundtrip_verified", "precision": precision,
|
| 233 |
+
"end_to_end_verification": "Requires separate manifest-bound equivalence.json (MTP=0) or equivalence-mtp.json (MTP=1); serialization alone is not validation",
|
| 234 |
+
"profile": args.profile, "files": files, "tensors": tensors,
|
| 235 |
+
"roundtrip": {"all_passed": True, "native_matrices": quantized, "lossless_tensors": counter - quantized,
|
| 236 |
+
"preserved_original_fp32_tensors": fp32,
|
| 237 |
+
"preserved_original_mtp_fp32_tensors": mtp_fp32},
|
| 238 |
+
"source": {"original_index_sha256": digest(index_path), "shards": source_shards,
|
| 239 |
+
"declared_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a"},
|
| 240 |
+
"toolchain": {"python": platform.python_version(), "torch": torch.__version__,
|
| 241 |
+
"ttnn": getattr(ttnn, "__version__", None), "container_image": args.container_image,
|
| 242 |
+
"tt_metal_commit": "de59f8a658b1ceafd230c8266026b1a72bb198d7"},
|
| 243 |
+
"load_contract": {"weights": "native dump -> CPU TT -> BF16 torch [out,in] -> runtime transpose -> requantize",
|
| 244 |
+
"nonlinear": "Lossless original tensors, preserving FP32 until normal runtime conversion",
|
| 245 |
+
"mtp": "Lossless original mtp.* tensors; existing Qwen36MTP uses target layer-0 precision policy with one speculative token",
|
| 246 |
+
"gdn_derived": "Runtime rebuilds AB and QKVABZ from quant-rounded components, then requantizes; no derived tensor stored",
|
| 247 |
+
"supported_runtime": "single-device Qwen36 text with optional MTP-1 at manifest precision, generic or explicitly dtype-gated packed families, only under verified runtime sources/environment",
|
| 248 |
+
"risks": ["Block exponent grouping is orientation/layout dependent", "Component idempotence does not prove GDN derived or full-model parity", "Every generic/packed configuration requires its own equivalence evidence; TP is excluded"],
|
| 249 |
+
"required_evidence": "native_checkpoint.py CHECKPOINT --baseline ORIGINAL_AT_SAME_PRECISION --restored NATIVE_RELOAD_RUN writes equivalence.json for recorded MTP=0 or equivalence-mtp.json for recorded MTP=1 only after exact full-logit comparison; actual speculation requires separate live-cycle evidence"},
|
| 250 |
+
"created_at": datetime.datetime.now(datetime.timezone.utc).isoformat()}
|
| 251 |
+
save_json(output / MANIFEST, manifest)
|
| 252 |
+
print(json.dumps({"manifest": str(output / MANIFEST), "tensors": counter, "native_matrices": quantized,
|
| 253 |
+
"artifact_bytes": sum(info["bytes"] for info in files.values()), "end_to_end_verified": False}), flush=True)
|
| 254 |
+
|
| 255 |
+
|
| 256 |
+
def main():
|
| 257 |
+
from tt_eval import PROFILES
|
| 258 |
+
|
| 259 |
+
parser = argparse.ArgumentParser(description=__doc__)
|
| 260 |
+
parser.add_argument("--weights", type=Path, required=True)
|
| 261 |
+
parser.add_argument("--output", type=Path, required=True)
|
| 262 |
+
parser.add_argument("--profile", choices=tuple(PROFILES), default="current-bfp4")
|
| 263 |
+
parser.add_argument("--overrides", type=Path)
|
| 264 |
+
parser.add_argument("--importance", type=Path)
|
| 265 |
+
parser.add_argument("--cpu-threads", type=int, default=8)
|
| 266 |
+
parser.add_argument("--container-image", required=True, help="Exact image identity reported by docker image inspect")
|
| 267 |
+
parser.add_argument("--device-ownership-confirmed", action="store_true")
|
| 268 |
+
args = parser.parse_args()
|
| 269 |
+
if not args.device_ownership_confirmed:
|
| 270 |
+
parser.error("TTNN host conversion requires exclusive device ownership; stop other P150 workloads first")
|
| 271 |
+
if args.cpu_threads < 1:
|
| 272 |
+
parser.error("--cpu-threads must be positive")
|
| 273 |
+
export(args)
|
| 274 |
+
|
| 275 |
+
|
| 276 |
+
if __name__ == "__main__":
|
| 277 |
+
main()
|
calibration/importance.npz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ca51deb3c0f6b096bea1ff2312b1bab4a4a8fa083ba0e03b282ec4f8ef303779
|
| 3 |
+
size 10427338
|
calibration/instruction-calibration.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
calibration/metadata.json
ADDED
|
@@ -0,0 +1,1337 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"arguments": {
|
| 3 |
+
"accumulation_chunk_rows": 128,
|
| 4 |
+
"checkpoint_every": 32,
|
| 5 |
+
"input": "/work/instruction-calibration.jsonl",
|
| 6 |
+
"interop_threads": 1,
|
| 7 |
+
"max_tokens": 2048,
|
| 8 |
+
"model": "/model",
|
| 9 |
+
"output": "/work/importance",
|
| 10 |
+
"revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 11 |
+
"sample_rows": 8,
|
| 12 |
+
"threads": 16
|
| 13 |
+
},
|
| 14 |
+
"completed_input_tokens": 300941,
|
| 15 |
+
"completed_records": 614,
|
| 16 |
+
"config_sha256": "d0883072e01861ed0b2d47be3c16c36a8e81c224c7ffaa310c6558fb3f932b05",
|
| 17 |
+
"elapsed_seconds": 2298.5781133160344,
|
| 18 |
+
"exclusions": "Only torch.nn.Linear input channels. Embedding lookup, GDN convolution, norms and A_log are not diagonal-linear importance entries; original reference weights remain unchanged.",
|
| 19 |
+
"importance_sha256": "ca51deb3c0f6b096bea1ff2312b1bab4a4a8fa083ba0e03b282ec4f8ef303779",
|
| 20 |
+
"index_sha256": "26d3539b516be613f39563617cb9d33b3f83d401298125be392c80cefb8f7fe5",
|
| 21 |
+
"input_sha256": "c65c34c37633381a7322ff433a292f9e615d414bb834d1951b243ee16ecb310e",
|
| 22 |
+
"kind": "reference_cpu_linear_input_diagonal_second_moments",
|
| 23 |
+
"last_record_id": "instruct-multilingual-train-145250",
|
| 24 |
+
"loading_info": {
|
| 25 |
+
"error_msgs": [],
|
| 26 |
+
"mismatched_keys": [],
|
| 27 |
+
"missing_keys": [],
|
| 28 |
+
"unexpected_keys": []
|
| 29 |
+
},
|
| 30 |
+
"memory_policy": "One batch-size-1 complete record <= max-tokens; CPU BF16 reference with original FP32 GDN weights; use_cache=False; no vocabulary projection/logits; reduction temp <= accumulation-chunk-rows * largest_linear_input_features FP32; persistent float64 channel sums and optional sample-rows FP32 activation rows per module.",
|
| 31 |
+
"model_id": "Qwen/Qwen3.5-9B",
|
| 32 |
+
"model_path": "/model",
|
| 33 |
+
"model_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 34 |
+
"module_key_format": "Original checkpoint module name (strip .weight), keys <module>.sumsq float64[input_features], <module>.count int64 scalar; optional <module>.samples float32[rows,input_features] and <module>.sample_token_indices int64[rows]",
|
| 35 |
+
"modules": {
|
| 36 |
+
"lm_head": {
|
| 37 |
+
"count": 300941,
|
| 38 |
+
"input_features": 4096,
|
| 39 |
+
"sample_rows": 8
|
| 40 |
+
},
|
| 41 |
+
"model.language_model.layers.0.linear_attn.in_proj_a": {
|
| 42 |
+
"count": 300941,
|
| 43 |
+
"input_features": 4096,
|
| 44 |
+
"sample_rows": 8
|
| 45 |
+
},
|
| 46 |
+
"model.language_model.layers.0.linear_attn.in_proj_b": {
|
| 47 |
+
"count": 300941,
|
| 48 |
+
"input_features": 4096,
|
| 49 |
+
"sample_rows": 8
|
| 50 |
+
},
|
| 51 |
+
"model.language_model.layers.0.linear_attn.in_proj_qkv": {
|
| 52 |
+
"count": 300941,
|
| 53 |
+
"input_features": 4096,
|
| 54 |
+
"sample_rows": 8
|
| 55 |
+
},
|
| 56 |
+
"model.language_model.layers.0.linear_attn.in_proj_z": {
|
| 57 |
+
"count": 300941,
|
| 58 |
+
"input_features": 4096,
|
| 59 |
+
"sample_rows": 8
|
| 60 |
+
},
|
| 61 |
+
"model.language_model.layers.0.linear_attn.out_proj": {
|
| 62 |
+
"count": 300941,
|
| 63 |
+
"input_features": 4096,
|
| 64 |
+
"sample_rows": 8
|
| 65 |
+
},
|
| 66 |
+
"model.language_model.layers.0.mlp.down_proj": {
|
| 67 |
+
"count": 300941,
|
| 68 |
+
"input_features": 12288,
|
| 69 |
+
"sample_rows": 8
|
| 70 |
+
},
|
| 71 |
+
"model.language_model.layers.0.mlp.gate_proj": {
|
| 72 |
+
"count": 300941,
|
| 73 |
+
"input_features": 4096,
|
| 74 |
+
"sample_rows": 8
|
| 75 |
+
},
|
| 76 |
+
"model.language_model.layers.0.mlp.up_proj": {
|
| 77 |
+
"count": 300941,
|
| 78 |
+
"input_features": 4096,
|
| 79 |
+
"sample_rows": 8
|
| 80 |
+
},
|
| 81 |
+
"model.language_model.layers.1.linear_attn.in_proj_a": {
|
| 82 |
+
"count": 300941,
|
| 83 |
+
"input_features": 4096,
|
| 84 |
+
"sample_rows": 8
|
| 85 |
+
},
|
| 86 |
+
"model.language_model.layers.1.linear_attn.in_proj_b": {
|
| 87 |
+
"count": 300941,
|
| 88 |
+
"input_features": 4096,
|
| 89 |
+
"sample_rows": 8
|
| 90 |
+
},
|
| 91 |
+
"model.language_model.layers.1.linear_attn.in_proj_qkv": {
|
| 92 |
+
"count": 300941,
|
| 93 |
+
"input_features": 4096,
|
| 94 |
+
"sample_rows": 8
|
| 95 |
+
},
|
| 96 |
+
"model.language_model.layers.1.linear_attn.in_proj_z": {
|
| 97 |
+
"count": 300941,
|
| 98 |
+
"input_features": 4096,
|
| 99 |
+
"sample_rows": 8
|
| 100 |
+
},
|
| 101 |
+
"model.language_model.layers.1.linear_attn.out_proj": {
|
| 102 |
+
"count": 300941,
|
| 103 |
+
"input_features": 4096,
|
| 104 |
+
"sample_rows": 8
|
| 105 |
+
},
|
| 106 |
+
"model.language_model.layers.1.mlp.down_proj": {
|
| 107 |
+
"count": 300941,
|
| 108 |
+
"input_features": 12288,
|
| 109 |
+
"sample_rows": 8
|
| 110 |
+
},
|
| 111 |
+
"model.language_model.layers.1.mlp.gate_proj": {
|
| 112 |
+
"count": 300941,
|
| 113 |
+
"input_features": 4096,
|
| 114 |
+
"sample_rows": 8
|
| 115 |
+
},
|
| 116 |
+
"model.language_model.layers.1.mlp.up_proj": {
|
| 117 |
+
"count": 300941,
|
| 118 |
+
"input_features": 4096,
|
| 119 |
+
"sample_rows": 8
|
| 120 |
+
},
|
| 121 |
+
"model.language_model.layers.10.linear_attn.in_proj_a": {
|
| 122 |
+
"count": 300941,
|
| 123 |
+
"input_features": 4096,
|
| 124 |
+
"sample_rows": 8
|
| 125 |
+
},
|
| 126 |
+
"model.language_model.layers.10.linear_attn.in_proj_b": {
|
| 127 |
+
"count": 300941,
|
| 128 |
+
"input_features": 4096,
|
| 129 |
+
"sample_rows": 8
|
| 130 |
+
},
|
| 131 |
+
"model.language_model.layers.10.linear_attn.in_proj_qkv": {
|
| 132 |
+
"count": 300941,
|
| 133 |
+
"input_features": 4096,
|
| 134 |
+
"sample_rows": 8
|
| 135 |
+
},
|
| 136 |
+
"model.language_model.layers.10.linear_attn.in_proj_z": {
|
| 137 |
+
"count": 300941,
|
| 138 |
+
"input_features": 4096,
|
| 139 |
+
"sample_rows": 8
|
| 140 |
+
},
|
| 141 |
+
"model.language_model.layers.10.linear_attn.out_proj": {
|
| 142 |
+
"count": 300941,
|
| 143 |
+
"input_features": 4096,
|
| 144 |
+
"sample_rows": 8
|
| 145 |
+
},
|
| 146 |
+
"model.language_model.layers.10.mlp.down_proj": {
|
| 147 |
+
"count": 300941,
|
| 148 |
+
"input_features": 12288,
|
| 149 |
+
"sample_rows": 8
|
| 150 |
+
},
|
| 151 |
+
"model.language_model.layers.10.mlp.gate_proj": {
|
| 152 |
+
"count": 300941,
|
| 153 |
+
"input_features": 4096,
|
| 154 |
+
"sample_rows": 8
|
| 155 |
+
},
|
| 156 |
+
"model.language_model.layers.10.mlp.up_proj": {
|
| 157 |
+
"count": 300941,
|
| 158 |
+
"input_features": 4096,
|
| 159 |
+
"sample_rows": 8
|
| 160 |
+
},
|
| 161 |
+
"model.language_model.layers.11.mlp.down_proj": {
|
| 162 |
+
"count": 300941,
|
| 163 |
+
"input_features": 12288,
|
| 164 |
+
"sample_rows": 8
|
| 165 |
+
},
|
| 166 |
+
"model.language_model.layers.11.mlp.gate_proj": {
|
| 167 |
+
"count": 300941,
|
| 168 |
+
"input_features": 4096,
|
| 169 |
+
"sample_rows": 8
|
| 170 |
+
},
|
| 171 |
+
"model.language_model.layers.11.mlp.up_proj": {
|
| 172 |
+
"count": 300941,
|
| 173 |
+
"input_features": 4096,
|
| 174 |
+
"sample_rows": 8
|
| 175 |
+
},
|
| 176 |
+
"model.language_model.layers.11.self_attn.k_proj": {
|
| 177 |
+
"count": 300941,
|
| 178 |
+
"input_features": 4096,
|
| 179 |
+
"sample_rows": 8
|
| 180 |
+
},
|
| 181 |
+
"model.language_model.layers.11.self_attn.o_proj": {
|
| 182 |
+
"count": 300941,
|
| 183 |
+
"input_features": 4096,
|
| 184 |
+
"sample_rows": 8
|
| 185 |
+
},
|
| 186 |
+
"model.language_model.layers.11.self_attn.q_proj": {
|
| 187 |
+
"count": 300941,
|
| 188 |
+
"input_features": 4096,
|
| 189 |
+
"sample_rows": 8
|
| 190 |
+
},
|
| 191 |
+
"model.language_model.layers.11.self_attn.v_proj": {
|
| 192 |
+
"count": 300941,
|
| 193 |
+
"input_features": 4096,
|
| 194 |
+
"sample_rows": 8
|
| 195 |
+
},
|
| 196 |
+
"model.language_model.layers.12.linear_attn.in_proj_a": {
|
| 197 |
+
"count": 300941,
|
| 198 |
+
"input_features": 4096,
|
| 199 |
+
"sample_rows": 8
|
| 200 |
+
},
|
| 201 |
+
"model.language_model.layers.12.linear_attn.in_proj_b": {
|
| 202 |
+
"count": 300941,
|
| 203 |
+
"input_features": 4096,
|
| 204 |
+
"sample_rows": 8
|
| 205 |
+
},
|
| 206 |
+
"model.language_model.layers.12.linear_attn.in_proj_qkv": {
|
| 207 |
+
"count": 300941,
|
| 208 |
+
"input_features": 4096,
|
| 209 |
+
"sample_rows": 8
|
| 210 |
+
},
|
| 211 |
+
"model.language_model.layers.12.linear_attn.in_proj_z": {
|
| 212 |
+
"count": 300941,
|
| 213 |
+
"input_features": 4096,
|
| 214 |
+
"sample_rows": 8
|
| 215 |
+
},
|
| 216 |
+
"model.language_model.layers.12.linear_attn.out_proj": {
|
| 217 |
+
"count": 300941,
|
| 218 |
+
"input_features": 4096,
|
| 219 |
+
"sample_rows": 8
|
| 220 |
+
},
|
| 221 |
+
"model.language_model.layers.12.mlp.down_proj": {
|
| 222 |
+
"count": 300941,
|
| 223 |
+
"input_features": 12288,
|
| 224 |
+
"sample_rows": 8
|
| 225 |
+
},
|
| 226 |
+
"model.language_model.layers.12.mlp.gate_proj": {
|
| 227 |
+
"count": 300941,
|
| 228 |
+
"input_features": 4096,
|
| 229 |
+
"sample_rows": 8
|
| 230 |
+
},
|
| 231 |
+
"model.language_model.layers.12.mlp.up_proj": {
|
| 232 |
+
"count": 300941,
|
| 233 |
+
"input_features": 4096,
|
| 234 |
+
"sample_rows": 8
|
| 235 |
+
},
|
| 236 |
+
"model.language_model.layers.13.linear_attn.in_proj_a": {
|
| 237 |
+
"count": 300941,
|
| 238 |
+
"input_features": 4096,
|
| 239 |
+
"sample_rows": 8
|
| 240 |
+
},
|
| 241 |
+
"model.language_model.layers.13.linear_attn.in_proj_b": {
|
| 242 |
+
"count": 300941,
|
| 243 |
+
"input_features": 4096,
|
| 244 |
+
"sample_rows": 8
|
| 245 |
+
},
|
| 246 |
+
"model.language_model.layers.13.linear_attn.in_proj_qkv": {
|
| 247 |
+
"count": 300941,
|
| 248 |
+
"input_features": 4096,
|
| 249 |
+
"sample_rows": 8
|
| 250 |
+
},
|
| 251 |
+
"model.language_model.layers.13.linear_attn.in_proj_z": {
|
| 252 |
+
"count": 300941,
|
| 253 |
+
"input_features": 4096,
|
| 254 |
+
"sample_rows": 8
|
| 255 |
+
},
|
| 256 |
+
"model.language_model.layers.13.linear_attn.out_proj": {
|
| 257 |
+
"count": 300941,
|
| 258 |
+
"input_features": 4096,
|
| 259 |
+
"sample_rows": 8
|
| 260 |
+
},
|
| 261 |
+
"model.language_model.layers.13.mlp.down_proj": {
|
| 262 |
+
"count": 300941,
|
| 263 |
+
"input_features": 12288,
|
| 264 |
+
"sample_rows": 8
|
| 265 |
+
},
|
| 266 |
+
"model.language_model.layers.13.mlp.gate_proj": {
|
| 267 |
+
"count": 300941,
|
| 268 |
+
"input_features": 4096,
|
| 269 |
+
"sample_rows": 8
|
| 270 |
+
},
|
| 271 |
+
"model.language_model.layers.13.mlp.up_proj": {
|
| 272 |
+
"count": 300941,
|
| 273 |
+
"input_features": 4096,
|
| 274 |
+
"sample_rows": 8
|
| 275 |
+
},
|
| 276 |
+
"model.language_model.layers.14.linear_attn.in_proj_a": {
|
| 277 |
+
"count": 300941,
|
| 278 |
+
"input_features": 4096,
|
| 279 |
+
"sample_rows": 8
|
| 280 |
+
},
|
| 281 |
+
"model.language_model.layers.14.linear_attn.in_proj_b": {
|
| 282 |
+
"count": 300941,
|
| 283 |
+
"input_features": 4096,
|
| 284 |
+
"sample_rows": 8
|
| 285 |
+
},
|
| 286 |
+
"model.language_model.layers.14.linear_attn.in_proj_qkv": {
|
| 287 |
+
"count": 300941,
|
| 288 |
+
"input_features": 4096,
|
| 289 |
+
"sample_rows": 8
|
| 290 |
+
},
|
| 291 |
+
"model.language_model.layers.14.linear_attn.in_proj_z": {
|
| 292 |
+
"count": 300941,
|
| 293 |
+
"input_features": 4096,
|
| 294 |
+
"sample_rows": 8
|
| 295 |
+
},
|
| 296 |
+
"model.language_model.layers.14.linear_attn.out_proj": {
|
| 297 |
+
"count": 300941,
|
| 298 |
+
"input_features": 4096,
|
| 299 |
+
"sample_rows": 8
|
| 300 |
+
},
|
| 301 |
+
"model.language_model.layers.14.mlp.down_proj": {
|
| 302 |
+
"count": 300941,
|
| 303 |
+
"input_features": 12288,
|
| 304 |
+
"sample_rows": 8
|
| 305 |
+
},
|
| 306 |
+
"model.language_model.layers.14.mlp.gate_proj": {
|
| 307 |
+
"count": 300941,
|
| 308 |
+
"input_features": 4096,
|
| 309 |
+
"sample_rows": 8
|
| 310 |
+
},
|
| 311 |
+
"model.language_model.layers.14.mlp.up_proj": {
|
| 312 |
+
"count": 300941,
|
| 313 |
+
"input_features": 4096,
|
| 314 |
+
"sample_rows": 8
|
| 315 |
+
},
|
| 316 |
+
"model.language_model.layers.15.mlp.down_proj": {
|
| 317 |
+
"count": 300941,
|
| 318 |
+
"input_features": 12288,
|
| 319 |
+
"sample_rows": 8
|
| 320 |
+
},
|
| 321 |
+
"model.language_model.layers.15.mlp.gate_proj": {
|
| 322 |
+
"count": 300941,
|
| 323 |
+
"input_features": 4096,
|
| 324 |
+
"sample_rows": 8
|
| 325 |
+
},
|
| 326 |
+
"model.language_model.layers.15.mlp.up_proj": {
|
| 327 |
+
"count": 300941,
|
| 328 |
+
"input_features": 4096,
|
| 329 |
+
"sample_rows": 8
|
| 330 |
+
},
|
| 331 |
+
"model.language_model.layers.15.self_attn.k_proj": {
|
| 332 |
+
"count": 300941,
|
| 333 |
+
"input_features": 4096,
|
| 334 |
+
"sample_rows": 8
|
| 335 |
+
},
|
| 336 |
+
"model.language_model.layers.15.self_attn.o_proj": {
|
| 337 |
+
"count": 300941,
|
| 338 |
+
"input_features": 4096,
|
| 339 |
+
"sample_rows": 8
|
| 340 |
+
},
|
| 341 |
+
"model.language_model.layers.15.self_attn.q_proj": {
|
| 342 |
+
"count": 300941,
|
| 343 |
+
"input_features": 4096,
|
| 344 |
+
"sample_rows": 8
|
| 345 |
+
},
|
| 346 |
+
"model.language_model.layers.15.self_attn.v_proj": {
|
| 347 |
+
"count": 300941,
|
| 348 |
+
"input_features": 4096,
|
| 349 |
+
"sample_rows": 8
|
| 350 |
+
},
|
| 351 |
+
"model.language_model.layers.16.linear_attn.in_proj_a": {
|
| 352 |
+
"count": 300941,
|
| 353 |
+
"input_features": 4096,
|
| 354 |
+
"sample_rows": 8
|
| 355 |
+
},
|
| 356 |
+
"model.language_model.layers.16.linear_attn.in_proj_b": {
|
| 357 |
+
"count": 300941,
|
| 358 |
+
"input_features": 4096,
|
| 359 |
+
"sample_rows": 8
|
| 360 |
+
},
|
| 361 |
+
"model.language_model.layers.16.linear_attn.in_proj_qkv": {
|
| 362 |
+
"count": 300941,
|
| 363 |
+
"input_features": 4096,
|
| 364 |
+
"sample_rows": 8
|
| 365 |
+
},
|
| 366 |
+
"model.language_model.layers.16.linear_attn.in_proj_z": {
|
| 367 |
+
"count": 300941,
|
| 368 |
+
"input_features": 4096,
|
| 369 |
+
"sample_rows": 8
|
| 370 |
+
},
|
| 371 |
+
"model.language_model.layers.16.linear_attn.out_proj": {
|
| 372 |
+
"count": 300941,
|
| 373 |
+
"input_features": 4096,
|
| 374 |
+
"sample_rows": 8
|
| 375 |
+
},
|
| 376 |
+
"model.language_model.layers.16.mlp.down_proj": {
|
| 377 |
+
"count": 300941,
|
| 378 |
+
"input_features": 12288,
|
| 379 |
+
"sample_rows": 8
|
| 380 |
+
},
|
| 381 |
+
"model.language_model.layers.16.mlp.gate_proj": {
|
| 382 |
+
"count": 300941,
|
| 383 |
+
"input_features": 4096,
|
| 384 |
+
"sample_rows": 8
|
| 385 |
+
},
|
| 386 |
+
"model.language_model.layers.16.mlp.up_proj": {
|
| 387 |
+
"count": 300941,
|
| 388 |
+
"input_features": 4096,
|
| 389 |
+
"sample_rows": 8
|
| 390 |
+
},
|
| 391 |
+
"model.language_model.layers.17.linear_attn.in_proj_a": {
|
| 392 |
+
"count": 300941,
|
| 393 |
+
"input_features": 4096,
|
| 394 |
+
"sample_rows": 8
|
| 395 |
+
},
|
| 396 |
+
"model.language_model.layers.17.linear_attn.in_proj_b": {
|
| 397 |
+
"count": 300941,
|
| 398 |
+
"input_features": 4096,
|
| 399 |
+
"sample_rows": 8
|
| 400 |
+
},
|
| 401 |
+
"model.language_model.layers.17.linear_attn.in_proj_qkv": {
|
| 402 |
+
"count": 300941,
|
| 403 |
+
"input_features": 4096,
|
| 404 |
+
"sample_rows": 8
|
| 405 |
+
},
|
| 406 |
+
"model.language_model.layers.17.linear_attn.in_proj_z": {
|
| 407 |
+
"count": 300941,
|
| 408 |
+
"input_features": 4096,
|
| 409 |
+
"sample_rows": 8
|
| 410 |
+
},
|
| 411 |
+
"model.language_model.layers.17.linear_attn.out_proj": {
|
| 412 |
+
"count": 300941,
|
| 413 |
+
"input_features": 4096,
|
| 414 |
+
"sample_rows": 8
|
| 415 |
+
},
|
| 416 |
+
"model.language_model.layers.17.mlp.down_proj": {
|
| 417 |
+
"count": 300941,
|
| 418 |
+
"input_features": 12288,
|
| 419 |
+
"sample_rows": 8
|
| 420 |
+
},
|
| 421 |
+
"model.language_model.layers.17.mlp.gate_proj": {
|
| 422 |
+
"count": 300941,
|
| 423 |
+
"input_features": 4096,
|
| 424 |
+
"sample_rows": 8
|
| 425 |
+
},
|
| 426 |
+
"model.language_model.layers.17.mlp.up_proj": {
|
| 427 |
+
"count": 300941,
|
| 428 |
+
"input_features": 4096,
|
| 429 |
+
"sample_rows": 8
|
| 430 |
+
},
|
| 431 |
+
"model.language_model.layers.18.linear_attn.in_proj_a": {
|
| 432 |
+
"count": 300941,
|
| 433 |
+
"input_features": 4096,
|
| 434 |
+
"sample_rows": 8
|
| 435 |
+
},
|
| 436 |
+
"model.language_model.layers.18.linear_attn.in_proj_b": {
|
| 437 |
+
"count": 300941,
|
| 438 |
+
"input_features": 4096,
|
| 439 |
+
"sample_rows": 8
|
| 440 |
+
},
|
| 441 |
+
"model.language_model.layers.18.linear_attn.in_proj_qkv": {
|
| 442 |
+
"count": 300941,
|
| 443 |
+
"input_features": 4096,
|
| 444 |
+
"sample_rows": 8
|
| 445 |
+
},
|
| 446 |
+
"model.language_model.layers.18.linear_attn.in_proj_z": {
|
| 447 |
+
"count": 300941,
|
| 448 |
+
"input_features": 4096,
|
| 449 |
+
"sample_rows": 8
|
| 450 |
+
},
|
| 451 |
+
"model.language_model.layers.18.linear_attn.out_proj": {
|
| 452 |
+
"count": 300941,
|
| 453 |
+
"input_features": 4096,
|
| 454 |
+
"sample_rows": 8
|
| 455 |
+
},
|
| 456 |
+
"model.language_model.layers.18.mlp.down_proj": {
|
| 457 |
+
"count": 300941,
|
| 458 |
+
"input_features": 12288,
|
| 459 |
+
"sample_rows": 8
|
| 460 |
+
},
|
| 461 |
+
"model.language_model.layers.18.mlp.gate_proj": {
|
| 462 |
+
"count": 300941,
|
| 463 |
+
"input_features": 4096,
|
| 464 |
+
"sample_rows": 8
|
| 465 |
+
},
|
| 466 |
+
"model.language_model.layers.18.mlp.up_proj": {
|
| 467 |
+
"count": 300941,
|
| 468 |
+
"input_features": 4096,
|
| 469 |
+
"sample_rows": 8
|
| 470 |
+
},
|
| 471 |
+
"model.language_model.layers.19.mlp.down_proj": {
|
| 472 |
+
"count": 300941,
|
| 473 |
+
"input_features": 12288,
|
| 474 |
+
"sample_rows": 8
|
| 475 |
+
},
|
| 476 |
+
"model.language_model.layers.19.mlp.gate_proj": {
|
| 477 |
+
"count": 300941,
|
| 478 |
+
"input_features": 4096,
|
| 479 |
+
"sample_rows": 8
|
| 480 |
+
},
|
| 481 |
+
"model.language_model.layers.19.mlp.up_proj": {
|
| 482 |
+
"count": 300941,
|
| 483 |
+
"input_features": 4096,
|
| 484 |
+
"sample_rows": 8
|
| 485 |
+
},
|
| 486 |
+
"model.language_model.layers.19.self_attn.k_proj": {
|
| 487 |
+
"count": 300941,
|
| 488 |
+
"input_features": 4096,
|
| 489 |
+
"sample_rows": 8
|
| 490 |
+
},
|
| 491 |
+
"model.language_model.layers.19.self_attn.o_proj": {
|
| 492 |
+
"count": 300941,
|
| 493 |
+
"input_features": 4096,
|
| 494 |
+
"sample_rows": 8
|
| 495 |
+
},
|
| 496 |
+
"model.language_model.layers.19.self_attn.q_proj": {
|
| 497 |
+
"count": 300941,
|
| 498 |
+
"input_features": 4096,
|
| 499 |
+
"sample_rows": 8
|
| 500 |
+
},
|
| 501 |
+
"model.language_model.layers.19.self_attn.v_proj": {
|
| 502 |
+
"count": 300941,
|
| 503 |
+
"input_features": 4096,
|
| 504 |
+
"sample_rows": 8
|
| 505 |
+
},
|
| 506 |
+
"model.language_model.layers.2.linear_attn.in_proj_a": {
|
| 507 |
+
"count": 300941,
|
| 508 |
+
"input_features": 4096,
|
| 509 |
+
"sample_rows": 8
|
| 510 |
+
},
|
| 511 |
+
"model.language_model.layers.2.linear_attn.in_proj_b": {
|
| 512 |
+
"count": 300941,
|
| 513 |
+
"input_features": 4096,
|
| 514 |
+
"sample_rows": 8
|
| 515 |
+
},
|
| 516 |
+
"model.language_model.layers.2.linear_attn.in_proj_qkv": {
|
| 517 |
+
"count": 300941,
|
| 518 |
+
"input_features": 4096,
|
| 519 |
+
"sample_rows": 8
|
| 520 |
+
},
|
| 521 |
+
"model.language_model.layers.2.linear_attn.in_proj_z": {
|
| 522 |
+
"count": 300941,
|
| 523 |
+
"input_features": 4096,
|
| 524 |
+
"sample_rows": 8
|
| 525 |
+
},
|
| 526 |
+
"model.language_model.layers.2.linear_attn.out_proj": {
|
| 527 |
+
"count": 300941,
|
| 528 |
+
"input_features": 4096,
|
| 529 |
+
"sample_rows": 8
|
| 530 |
+
},
|
| 531 |
+
"model.language_model.layers.2.mlp.down_proj": {
|
| 532 |
+
"count": 300941,
|
| 533 |
+
"input_features": 12288,
|
| 534 |
+
"sample_rows": 8
|
| 535 |
+
},
|
| 536 |
+
"model.language_model.layers.2.mlp.gate_proj": {
|
| 537 |
+
"count": 300941,
|
| 538 |
+
"input_features": 4096,
|
| 539 |
+
"sample_rows": 8
|
| 540 |
+
},
|
| 541 |
+
"model.language_model.layers.2.mlp.up_proj": {
|
| 542 |
+
"count": 300941,
|
| 543 |
+
"input_features": 4096,
|
| 544 |
+
"sample_rows": 8
|
| 545 |
+
},
|
| 546 |
+
"model.language_model.layers.20.linear_attn.in_proj_a": {
|
| 547 |
+
"count": 300941,
|
| 548 |
+
"input_features": 4096,
|
| 549 |
+
"sample_rows": 8
|
| 550 |
+
},
|
| 551 |
+
"model.language_model.layers.20.linear_attn.in_proj_b": {
|
| 552 |
+
"count": 300941,
|
| 553 |
+
"input_features": 4096,
|
| 554 |
+
"sample_rows": 8
|
| 555 |
+
},
|
| 556 |
+
"model.language_model.layers.20.linear_attn.in_proj_qkv": {
|
| 557 |
+
"count": 300941,
|
| 558 |
+
"input_features": 4096,
|
| 559 |
+
"sample_rows": 8
|
| 560 |
+
},
|
| 561 |
+
"model.language_model.layers.20.linear_attn.in_proj_z": {
|
| 562 |
+
"count": 300941,
|
| 563 |
+
"input_features": 4096,
|
| 564 |
+
"sample_rows": 8
|
| 565 |
+
},
|
| 566 |
+
"model.language_model.layers.20.linear_attn.out_proj": {
|
| 567 |
+
"count": 300941,
|
| 568 |
+
"input_features": 4096,
|
| 569 |
+
"sample_rows": 8
|
| 570 |
+
},
|
| 571 |
+
"model.language_model.layers.20.mlp.down_proj": {
|
| 572 |
+
"count": 300941,
|
| 573 |
+
"input_features": 12288,
|
| 574 |
+
"sample_rows": 8
|
| 575 |
+
},
|
| 576 |
+
"model.language_model.layers.20.mlp.gate_proj": {
|
| 577 |
+
"count": 300941,
|
| 578 |
+
"input_features": 4096,
|
| 579 |
+
"sample_rows": 8
|
| 580 |
+
},
|
| 581 |
+
"model.language_model.layers.20.mlp.up_proj": {
|
| 582 |
+
"count": 300941,
|
| 583 |
+
"input_features": 4096,
|
| 584 |
+
"sample_rows": 8
|
| 585 |
+
},
|
| 586 |
+
"model.language_model.layers.21.linear_attn.in_proj_a": {
|
| 587 |
+
"count": 300941,
|
| 588 |
+
"input_features": 4096,
|
| 589 |
+
"sample_rows": 8
|
| 590 |
+
},
|
| 591 |
+
"model.language_model.layers.21.linear_attn.in_proj_b": {
|
| 592 |
+
"count": 300941,
|
| 593 |
+
"input_features": 4096,
|
| 594 |
+
"sample_rows": 8
|
| 595 |
+
},
|
| 596 |
+
"model.language_model.layers.21.linear_attn.in_proj_qkv": {
|
| 597 |
+
"count": 300941,
|
| 598 |
+
"input_features": 4096,
|
| 599 |
+
"sample_rows": 8
|
| 600 |
+
},
|
| 601 |
+
"model.language_model.layers.21.linear_attn.in_proj_z": {
|
| 602 |
+
"count": 300941,
|
| 603 |
+
"input_features": 4096,
|
| 604 |
+
"sample_rows": 8
|
| 605 |
+
},
|
| 606 |
+
"model.language_model.layers.21.linear_attn.out_proj": {
|
| 607 |
+
"count": 300941,
|
| 608 |
+
"input_features": 4096,
|
| 609 |
+
"sample_rows": 8
|
| 610 |
+
},
|
| 611 |
+
"model.language_model.layers.21.mlp.down_proj": {
|
| 612 |
+
"count": 300941,
|
| 613 |
+
"input_features": 12288,
|
| 614 |
+
"sample_rows": 8
|
| 615 |
+
},
|
| 616 |
+
"model.language_model.layers.21.mlp.gate_proj": {
|
| 617 |
+
"count": 300941,
|
| 618 |
+
"input_features": 4096,
|
| 619 |
+
"sample_rows": 8
|
| 620 |
+
},
|
| 621 |
+
"model.language_model.layers.21.mlp.up_proj": {
|
| 622 |
+
"count": 300941,
|
| 623 |
+
"input_features": 4096,
|
| 624 |
+
"sample_rows": 8
|
| 625 |
+
},
|
| 626 |
+
"model.language_model.layers.22.linear_attn.in_proj_a": {
|
| 627 |
+
"count": 300941,
|
| 628 |
+
"input_features": 4096,
|
| 629 |
+
"sample_rows": 8
|
| 630 |
+
},
|
| 631 |
+
"model.language_model.layers.22.linear_attn.in_proj_b": {
|
| 632 |
+
"count": 300941,
|
| 633 |
+
"input_features": 4096,
|
| 634 |
+
"sample_rows": 8
|
| 635 |
+
},
|
| 636 |
+
"model.language_model.layers.22.linear_attn.in_proj_qkv": {
|
| 637 |
+
"count": 300941,
|
| 638 |
+
"input_features": 4096,
|
| 639 |
+
"sample_rows": 8
|
| 640 |
+
},
|
| 641 |
+
"model.language_model.layers.22.linear_attn.in_proj_z": {
|
| 642 |
+
"count": 300941,
|
| 643 |
+
"input_features": 4096,
|
| 644 |
+
"sample_rows": 8
|
| 645 |
+
},
|
| 646 |
+
"model.language_model.layers.22.linear_attn.out_proj": {
|
| 647 |
+
"count": 300941,
|
| 648 |
+
"input_features": 4096,
|
| 649 |
+
"sample_rows": 8
|
| 650 |
+
},
|
| 651 |
+
"model.language_model.layers.22.mlp.down_proj": {
|
| 652 |
+
"count": 300941,
|
| 653 |
+
"input_features": 12288,
|
| 654 |
+
"sample_rows": 8
|
| 655 |
+
},
|
| 656 |
+
"model.language_model.layers.22.mlp.gate_proj": {
|
| 657 |
+
"count": 300941,
|
| 658 |
+
"input_features": 4096,
|
| 659 |
+
"sample_rows": 8
|
| 660 |
+
},
|
| 661 |
+
"model.language_model.layers.22.mlp.up_proj": {
|
| 662 |
+
"count": 300941,
|
| 663 |
+
"input_features": 4096,
|
| 664 |
+
"sample_rows": 8
|
| 665 |
+
},
|
| 666 |
+
"model.language_model.layers.23.mlp.down_proj": {
|
| 667 |
+
"count": 300941,
|
| 668 |
+
"input_features": 12288,
|
| 669 |
+
"sample_rows": 8
|
| 670 |
+
},
|
| 671 |
+
"model.language_model.layers.23.mlp.gate_proj": {
|
| 672 |
+
"count": 300941,
|
| 673 |
+
"input_features": 4096,
|
| 674 |
+
"sample_rows": 8
|
| 675 |
+
},
|
| 676 |
+
"model.language_model.layers.23.mlp.up_proj": {
|
| 677 |
+
"count": 300941,
|
| 678 |
+
"input_features": 4096,
|
| 679 |
+
"sample_rows": 8
|
| 680 |
+
},
|
| 681 |
+
"model.language_model.layers.23.self_attn.k_proj": {
|
| 682 |
+
"count": 300941,
|
| 683 |
+
"input_features": 4096,
|
| 684 |
+
"sample_rows": 8
|
| 685 |
+
},
|
| 686 |
+
"model.language_model.layers.23.self_attn.o_proj": {
|
| 687 |
+
"count": 300941,
|
| 688 |
+
"input_features": 4096,
|
| 689 |
+
"sample_rows": 8
|
| 690 |
+
},
|
| 691 |
+
"model.language_model.layers.23.self_attn.q_proj": {
|
| 692 |
+
"count": 300941,
|
| 693 |
+
"input_features": 4096,
|
| 694 |
+
"sample_rows": 8
|
| 695 |
+
},
|
| 696 |
+
"model.language_model.layers.23.self_attn.v_proj": {
|
| 697 |
+
"count": 300941,
|
| 698 |
+
"input_features": 4096,
|
| 699 |
+
"sample_rows": 8
|
| 700 |
+
},
|
| 701 |
+
"model.language_model.layers.24.linear_attn.in_proj_a": {
|
| 702 |
+
"count": 300941,
|
| 703 |
+
"input_features": 4096,
|
| 704 |
+
"sample_rows": 8
|
| 705 |
+
},
|
| 706 |
+
"model.language_model.layers.24.linear_attn.in_proj_b": {
|
| 707 |
+
"count": 300941,
|
| 708 |
+
"input_features": 4096,
|
| 709 |
+
"sample_rows": 8
|
| 710 |
+
},
|
| 711 |
+
"model.language_model.layers.24.linear_attn.in_proj_qkv": {
|
| 712 |
+
"count": 300941,
|
| 713 |
+
"input_features": 4096,
|
| 714 |
+
"sample_rows": 8
|
| 715 |
+
},
|
| 716 |
+
"model.language_model.layers.24.linear_attn.in_proj_z": {
|
| 717 |
+
"count": 300941,
|
| 718 |
+
"input_features": 4096,
|
| 719 |
+
"sample_rows": 8
|
| 720 |
+
},
|
| 721 |
+
"model.language_model.layers.24.linear_attn.out_proj": {
|
| 722 |
+
"count": 300941,
|
| 723 |
+
"input_features": 4096,
|
| 724 |
+
"sample_rows": 8
|
| 725 |
+
},
|
| 726 |
+
"model.language_model.layers.24.mlp.down_proj": {
|
| 727 |
+
"count": 300941,
|
| 728 |
+
"input_features": 12288,
|
| 729 |
+
"sample_rows": 8
|
| 730 |
+
},
|
| 731 |
+
"model.language_model.layers.24.mlp.gate_proj": {
|
| 732 |
+
"count": 300941,
|
| 733 |
+
"input_features": 4096,
|
| 734 |
+
"sample_rows": 8
|
| 735 |
+
},
|
| 736 |
+
"model.language_model.layers.24.mlp.up_proj": {
|
| 737 |
+
"count": 300941,
|
| 738 |
+
"input_features": 4096,
|
| 739 |
+
"sample_rows": 8
|
| 740 |
+
},
|
| 741 |
+
"model.language_model.layers.25.linear_attn.in_proj_a": {
|
| 742 |
+
"count": 300941,
|
| 743 |
+
"input_features": 4096,
|
| 744 |
+
"sample_rows": 8
|
| 745 |
+
},
|
| 746 |
+
"model.language_model.layers.25.linear_attn.in_proj_b": {
|
| 747 |
+
"count": 300941,
|
| 748 |
+
"input_features": 4096,
|
| 749 |
+
"sample_rows": 8
|
| 750 |
+
},
|
| 751 |
+
"model.language_model.layers.25.linear_attn.in_proj_qkv": {
|
| 752 |
+
"count": 300941,
|
| 753 |
+
"input_features": 4096,
|
| 754 |
+
"sample_rows": 8
|
| 755 |
+
},
|
| 756 |
+
"model.language_model.layers.25.linear_attn.in_proj_z": {
|
| 757 |
+
"count": 300941,
|
| 758 |
+
"input_features": 4096,
|
| 759 |
+
"sample_rows": 8
|
| 760 |
+
},
|
| 761 |
+
"model.language_model.layers.25.linear_attn.out_proj": {
|
| 762 |
+
"count": 300941,
|
| 763 |
+
"input_features": 4096,
|
| 764 |
+
"sample_rows": 8
|
| 765 |
+
},
|
| 766 |
+
"model.language_model.layers.25.mlp.down_proj": {
|
| 767 |
+
"count": 300941,
|
| 768 |
+
"input_features": 12288,
|
| 769 |
+
"sample_rows": 8
|
| 770 |
+
},
|
| 771 |
+
"model.language_model.layers.25.mlp.gate_proj": {
|
| 772 |
+
"count": 300941,
|
| 773 |
+
"input_features": 4096,
|
| 774 |
+
"sample_rows": 8
|
| 775 |
+
},
|
| 776 |
+
"model.language_model.layers.25.mlp.up_proj": {
|
| 777 |
+
"count": 300941,
|
| 778 |
+
"input_features": 4096,
|
| 779 |
+
"sample_rows": 8
|
| 780 |
+
},
|
| 781 |
+
"model.language_model.layers.26.linear_attn.in_proj_a": {
|
| 782 |
+
"count": 300941,
|
| 783 |
+
"input_features": 4096,
|
| 784 |
+
"sample_rows": 8
|
| 785 |
+
},
|
| 786 |
+
"model.language_model.layers.26.linear_attn.in_proj_b": {
|
| 787 |
+
"count": 300941,
|
| 788 |
+
"input_features": 4096,
|
| 789 |
+
"sample_rows": 8
|
| 790 |
+
},
|
| 791 |
+
"model.language_model.layers.26.linear_attn.in_proj_qkv": {
|
| 792 |
+
"count": 300941,
|
| 793 |
+
"input_features": 4096,
|
| 794 |
+
"sample_rows": 8
|
| 795 |
+
},
|
| 796 |
+
"model.language_model.layers.26.linear_attn.in_proj_z": {
|
| 797 |
+
"count": 300941,
|
| 798 |
+
"input_features": 4096,
|
| 799 |
+
"sample_rows": 8
|
| 800 |
+
},
|
| 801 |
+
"model.language_model.layers.26.linear_attn.out_proj": {
|
| 802 |
+
"count": 300941,
|
| 803 |
+
"input_features": 4096,
|
| 804 |
+
"sample_rows": 8
|
| 805 |
+
},
|
| 806 |
+
"model.language_model.layers.26.mlp.down_proj": {
|
| 807 |
+
"count": 300941,
|
| 808 |
+
"input_features": 12288,
|
| 809 |
+
"sample_rows": 8
|
| 810 |
+
},
|
| 811 |
+
"model.language_model.layers.26.mlp.gate_proj": {
|
| 812 |
+
"count": 300941,
|
| 813 |
+
"input_features": 4096,
|
| 814 |
+
"sample_rows": 8
|
| 815 |
+
},
|
| 816 |
+
"model.language_model.layers.26.mlp.up_proj": {
|
| 817 |
+
"count": 300941,
|
| 818 |
+
"input_features": 4096,
|
| 819 |
+
"sample_rows": 8
|
| 820 |
+
},
|
| 821 |
+
"model.language_model.layers.27.mlp.down_proj": {
|
| 822 |
+
"count": 300941,
|
| 823 |
+
"input_features": 12288,
|
| 824 |
+
"sample_rows": 8
|
| 825 |
+
},
|
| 826 |
+
"model.language_model.layers.27.mlp.gate_proj": {
|
| 827 |
+
"count": 300941,
|
| 828 |
+
"input_features": 4096,
|
| 829 |
+
"sample_rows": 8
|
| 830 |
+
},
|
| 831 |
+
"model.language_model.layers.27.mlp.up_proj": {
|
| 832 |
+
"count": 300941,
|
| 833 |
+
"input_features": 4096,
|
| 834 |
+
"sample_rows": 8
|
| 835 |
+
},
|
| 836 |
+
"model.language_model.layers.27.self_attn.k_proj": {
|
| 837 |
+
"count": 300941,
|
| 838 |
+
"input_features": 4096,
|
| 839 |
+
"sample_rows": 8
|
| 840 |
+
},
|
| 841 |
+
"model.language_model.layers.27.self_attn.o_proj": {
|
| 842 |
+
"count": 300941,
|
| 843 |
+
"input_features": 4096,
|
| 844 |
+
"sample_rows": 8
|
| 845 |
+
},
|
| 846 |
+
"model.language_model.layers.27.self_attn.q_proj": {
|
| 847 |
+
"count": 300941,
|
| 848 |
+
"input_features": 4096,
|
| 849 |
+
"sample_rows": 8
|
| 850 |
+
},
|
| 851 |
+
"model.language_model.layers.27.self_attn.v_proj": {
|
| 852 |
+
"count": 300941,
|
| 853 |
+
"input_features": 4096,
|
| 854 |
+
"sample_rows": 8
|
| 855 |
+
},
|
| 856 |
+
"model.language_model.layers.28.linear_attn.in_proj_a": {
|
| 857 |
+
"count": 300941,
|
| 858 |
+
"input_features": 4096,
|
| 859 |
+
"sample_rows": 8
|
| 860 |
+
},
|
| 861 |
+
"model.language_model.layers.28.linear_attn.in_proj_b": {
|
| 862 |
+
"count": 300941,
|
| 863 |
+
"input_features": 4096,
|
| 864 |
+
"sample_rows": 8
|
| 865 |
+
},
|
| 866 |
+
"model.language_model.layers.28.linear_attn.in_proj_qkv": {
|
| 867 |
+
"count": 300941,
|
| 868 |
+
"input_features": 4096,
|
| 869 |
+
"sample_rows": 8
|
| 870 |
+
},
|
| 871 |
+
"model.language_model.layers.28.linear_attn.in_proj_z": {
|
| 872 |
+
"count": 300941,
|
| 873 |
+
"input_features": 4096,
|
| 874 |
+
"sample_rows": 8
|
| 875 |
+
},
|
| 876 |
+
"model.language_model.layers.28.linear_attn.out_proj": {
|
| 877 |
+
"count": 300941,
|
| 878 |
+
"input_features": 4096,
|
| 879 |
+
"sample_rows": 8
|
| 880 |
+
},
|
| 881 |
+
"model.language_model.layers.28.mlp.down_proj": {
|
| 882 |
+
"count": 300941,
|
| 883 |
+
"input_features": 12288,
|
| 884 |
+
"sample_rows": 8
|
| 885 |
+
},
|
| 886 |
+
"model.language_model.layers.28.mlp.gate_proj": {
|
| 887 |
+
"count": 300941,
|
| 888 |
+
"input_features": 4096,
|
| 889 |
+
"sample_rows": 8
|
| 890 |
+
},
|
| 891 |
+
"model.language_model.layers.28.mlp.up_proj": {
|
| 892 |
+
"count": 300941,
|
| 893 |
+
"input_features": 4096,
|
| 894 |
+
"sample_rows": 8
|
| 895 |
+
},
|
| 896 |
+
"model.language_model.layers.29.linear_attn.in_proj_a": {
|
| 897 |
+
"count": 300941,
|
| 898 |
+
"input_features": 4096,
|
| 899 |
+
"sample_rows": 8
|
| 900 |
+
},
|
| 901 |
+
"model.language_model.layers.29.linear_attn.in_proj_b": {
|
| 902 |
+
"count": 300941,
|
| 903 |
+
"input_features": 4096,
|
| 904 |
+
"sample_rows": 8
|
| 905 |
+
},
|
| 906 |
+
"model.language_model.layers.29.linear_attn.in_proj_qkv": {
|
| 907 |
+
"count": 300941,
|
| 908 |
+
"input_features": 4096,
|
| 909 |
+
"sample_rows": 8
|
| 910 |
+
},
|
| 911 |
+
"model.language_model.layers.29.linear_attn.in_proj_z": {
|
| 912 |
+
"count": 300941,
|
| 913 |
+
"input_features": 4096,
|
| 914 |
+
"sample_rows": 8
|
| 915 |
+
},
|
| 916 |
+
"model.language_model.layers.29.linear_attn.out_proj": {
|
| 917 |
+
"count": 300941,
|
| 918 |
+
"input_features": 4096,
|
| 919 |
+
"sample_rows": 8
|
| 920 |
+
},
|
| 921 |
+
"model.language_model.layers.29.mlp.down_proj": {
|
| 922 |
+
"count": 300941,
|
| 923 |
+
"input_features": 12288,
|
| 924 |
+
"sample_rows": 8
|
| 925 |
+
},
|
| 926 |
+
"model.language_model.layers.29.mlp.gate_proj": {
|
| 927 |
+
"count": 300941,
|
| 928 |
+
"input_features": 4096,
|
| 929 |
+
"sample_rows": 8
|
| 930 |
+
},
|
| 931 |
+
"model.language_model.layers.29.mlp.up_proj": {
|
| 932 |
+
"count": 300941,
|
| 933 |
+
"input_features": 4096,
|
| 934 |
+
"sample_rows": 8
|
| 935 |
+
},
|
| 936 |
+
"model.language_model.layers.3.mlp.down_proj": {
|
| 937 |
+
"count": 300941,
|
| 938 |
+
"input_features": 12288,
|
| 939 |
+
"sample_rows": 8
|
| 940 |
+
},
|
| 941 |
+
"model.language_model.layers.3.mlp.gate_proj": {
|
| 942 |
+
"count": 300941,
|
| 943 |
+
"input_features": 4096,
|
| 944 |
+
"sample_rows": 8
|
| 945 |
+
},
|
| 946 |
+
"model.language_model.layers.3.mlp.up_proj": {
|
| 947 |
+
"count": 300941,
|
| 948 |
+
"input_features": 4096,
|
| 949 |
+
"sample_rows": 8
|
| 950 |
+
},
|
| 951 |
+
"model.language_model.layers.3.self_attn.k_proj": {
|
| 952 |
+
"count": 300941,
|
| 953 |
+
"input_features": 4096,
|
| 954 |
+
"sample_rows": 8
|
| 955 |
+
},
|
| 956 |
+
"model.language_model.layers.3.self_attn.o_proj": {
|
| 957 |
+
"count": 300941,
|
| 958 |
+
"input_features": 4096,
|
| 959 |
+
"sample_rows": 8
|
| 960 |
+
},
|
| 961 |
+
"model.language_model.layers.3.self_attn.q_proj": {
|
| 962 |
+
"count": 300941,
|
| 963 |
+
"input_features": 4096,
|
| 964 |
+
"sample_rows": 8
|
| 965 |
+
},
|
| 966 |
+
"model.language_model.layers.3.self_attn.v_proj": {
|
| 967 |
+
"count": 300941,
|
| 968 |
+
"input_features": 4096,
|
| 969 |
+
"sample_rows": 8
|
| 970 |
+
},
|
| 971 |
+
"model.language_model.layers.30.linear_attn.in_proj_a": {
|
| 972 |
+
"count": 300941,
|
| 973 |
+
"input_features": 4096,
|
| 974 |
+
"sample_rows": 8
|
| 975 |
+
},
|
| 976 |
+
"model.language_model.layers.30.linear_attn.in_proj_b": {
|
| 977 |
+
"count": 300941,
|
| 978 |
+
"input_features": 4096,
|
| 979 |
+
"sample_rows": 8
|
| 980 |
+
},
|
| 981 |
+
"model.language_model.layers.30.linear_attn.in_proj_qkv": {
|
| 982 |
+
"count": 300941,
|
| 983 |
+
"input_features": 4096,
|
| 984 |
+
"sample_rows": 8
|
| 985 |
+
},
|
| 986 |
+
"model.language_model.layers.30.linear_attn.in_proj_z": {
|
| 987 |
+
"count": 300941,
|
| 988 |
+
"input_features": 4096,
|
| 989 |
+
"sample_rows": 8
|
| 990 |
+
},
|
| 991 |
+
"model.language_model.layers.30.linear_attn.out_proj": {
|
| 992 |
+
"count": 300941,
|
| 993 |
+
"input_features": 4096,
|
| 994 |
+
"sample_rows": 8
|
| 995 |
+
},
|
| 996 |
+
"model.language_model.layers.30.mlp.down_proj": {
|
| 997 |
+
"count": 300941,
|
| 998 |
+
"input_features": 12288,
|
| 999 |
+
"sample_rows": 8
|
| 1000 |
+
},
|
| 1001 |
+
"model.language_model.layers.30.mlp.gate_proj": {
|
| 1002 |
+
"count": 300941,
|
| 1003 |
+
"input_features": 4096,
|
| 1004 |
+
"sample_rows": 8
|
| 1005 |
+
},
|
| 1006 |
+
"model.language_model.layers.30.mlp.up_proj": {
|
| 1007 |
+
"count": 300941,
|
| 1008 |
+
"input_features": 4096,
|
| 1009 |
+
"sample_rows": 8
|
| 1010 |
+
},
|
| 1011 |
+
"model.language_model.layers.31.mlp.down_proj": {
|
| 1012 |
+
"count": 300941,
|
| 1013 |
+
"input_features": 12288,
|
| 1014 |
+
"sample_rows": 8
|
| 1015 |
+
},
|
| 1016 |
+
"model.language_model.layers.31.mlp.gate_proj": {
|
| 1017 |
+
"count": 300941,
|
| 1018 |
+
"input_features": 4096,
|
| 1019 |
+
"sample_rows": 8
|
| 1020 |
+
},
|
| 1021 |
+
"model.language_model.layers.31.mlp.up_proj": {
|
| 1022 |
+
"count": 300941,
|
| 1023 |
+
"input_features": 4096,
|
| 1024 |
+
"sample_rows": 8
|
| 1025 |
+
},
|
| 1026 |
+
"model.language_model.layers.31.self_attn.k_proj": {
|
| 1027 |
+
"count": 300941,
|
| 1028 |
+
"input_features": 4096,
|
| 1029 |
+
"sample_rows": 8
|
| 1030 |
+
},
|
| 1031 |
+
"model.language_model.layers.31.self_attn.o_proj": {
|
| 1032 |
+
"count": 300941,
|
| 1033 |
+
"input_features": 4096,
|
| 1034 |
+
"sample_rows": 8
|
| 1035 |
+
},
|
| 1036 |
+
"model.language_model.layers.31.self_attn.q_proj": {
|
| 1037 |
+
"count": 300941,
|
| 1038 |
+
"input_features": 4096,
|
| 1039 |
+
"sample_rows": 8
|
| 1040 |
+
},
|
| 1041 |
+
"model.language_model.layers.31.self_attn.v_proj": {
|
| 1042 |
+
"count": 300941,
|
| 1043 |
+
"input_features": 4096,
|
| 1044 |
+
"sample_rows": 8
|
| 1045 |
+
},
|
| 1046 |
+
"model.language_model.layers.4.linear_attn.in_proj_a": {
|
| 1047 |
+
"count": 300941,
|
| 1048 |
+
"input_features": 4096,
|
| 1049 |
+
"sample_rows": 8
|
| 1050 |
+
},
|
| 1051 |
+
"model.language_model.layers.4.linear_attn.in_proj_b": {
|
| 1052 |
+
"count": 300941,
|
| 1053 |
+
"input_features": 4096,
|
| 1054 |
+
"sample_rows": 8
|
| 1055 |
+
},
|
| 1056 |
+
"model.language_model.layers.4.linear_attn.in_proj_qkv": {
|
| 1057 |
+
"count": 300941,
|
| 1058 |
+
"input_features": 4096,
|
| 1059 |
+
"sample_rows": 8
|
| 1060 |
+
},
|
| 1061 |
+
"model.language_model.layers.4.linear_attn.in_proj_z": {
|
| 1062 |
+
"count": 300941,
|
| 1063 |
+
"input_features": 4096,
|
| 1064 |
+
"sample_rows": 8
|
| 1065 |
+
},
|
| 1066 |
+
"model.language_model.layers.4.linear_attn.out_proj": {
|
| 1067 |
+
"count": 300941,
|
| 1068 |
+
"input_features": 4096,
|
| 1069 |
+
"sample_rows": 8
|
| 1070 |
+
},
|
| 1071 |
+
"model.language_model.layers.4.mlp.down_proj": {
|
| 1072 |
+
"count": 300941,
|
| 1073 |
+
"input_features": 12288,
|
| 1074 |
+
"sample_rows": 8
|
| 1075 |
+
},
|
| 1076 |
+
"model.language_model.layers.4.mlp.gate_proj": {
|
| 1077 |
+
"count": 300941,
|
| 1078 |
+
"input_features": 4096,
|
| 1079 |
+
"sample_rows": 8
|
| 1080 |
+
},
|
| 1081 |
+
"model.language_model.layers.4.mlp.up_proj": {
|
| 1082 |
+
"count": 300941,
|
| 1083 |
+
"input_features": 4096,
|
| 1084 |
+
"sample_rows": 8
|
| 1085 |
+
},
|
| 1086 |
+
"model.language_model.layers.5.linear_attn.in_proj_a": {
|
| 1087 |
+
"count": 300941,
|
| 1088 |
+
"input_features": 4096,
|
| 1089 |
+
"sample_rows": 8
|
| 1090 |
+
},
|
| 1091 |
+
"model.language_model.layers.5.linear_attn.in_proj_b": {
|
| 1092 |
+
"count": 300941,
|
| 1093 |
+
"input_features": 4096,
|
| 1094 |
+
"sample_rows": 8
|
| 1095 |
+
},
|
| 1096 |
+
"model.language_model.layers.5.linear_attn.in_proj_qkv": {
|
| 1097 |
+
"count": 300941,
|
| 1098 |
+
"input_features": 4096,
|
| 1099 |
+
"sample_rows": 8
|
| 1100 |
+
},
|
| 1101 |
+
"model.language_model.layers.5.linear_attn.in_proj_z": {
|
| 1102 |
+
"count": 300941,
|
| 1103 |
+
"input_features": 4096,
|
| 1104 |
+
"sample_rows": 8
|
| 1105 |
+
},
|
| 1106 |
+
"model.language_model.layers.5.linear_attn.out_proj": {
|
| 1107 |
+
"count": 300941,
|
| 1108 |
+
"input_features": 4096,
|
| 1109 |
+
"sample_rows": 8
|
| 1110 |
+
},
|
| 1111 |
+
"model.language_model.layers.5.mlp.down_proj": {
|
| 1112 |
+
"count": 300941,
|
| 1113 |
+
"input_features": 12288,
|
| 1114 |
+
"sample_rows": 8
|
| 1115 |
+
},
|
| 1116 |
+
"model.language_model.layers.5.mlp.gate_proj": {
|
| 1117 |
+
"count": 300941,
|
| 1118 |
+
"input_features": 4096,
|
| 1119 |
+
"sample_rows": 8
|
| 1120 |
+
},
|
| 1121 |
+
"model.language_model.layers.5.mlp.up_proj": {
|
| 1122 |
+
"count": 300941,
|
| 1123 |
+
"input_features": 4096,
|
| 1124 |
+
"sample_rows": 8
|
| 1125 |
+
},
|
| 1126 |
+
"model.language_model.layers.6.linear_attn.in_proj_a": {
|
| 1127 |
+
"count": 300941,
|
| 1128 |
+
"input_features": 4096,
|
| 1129 |
+
"sample_rows": 8
|
| 1130 |
+
},
|
| 1131 |
+
"model.language_model.layers.6.linear_attn.in_proj_b": {
|
| 1132 |
+
"count": 300941,
|
| 1133 |
+
"input_features": 4096,
|
| 1134 |
+
"sample_rows": 8
|
| 1135 |
+
},
|
| 1136 |
+
"model.language_model.layers.6.linear_attn.in_proj_qkv": {
|
| 1137 |
+
"count": 300941,
|
| 1138 |
+
"input_features": 4096,
|
| 1139 |
+
"sample_rows": 8
|
| 1140 |
+
},
|
| 1141 |
+
"model.language_model.layers.6.linear_attn.in_proj_z": {
|
| 1142 |
+
"count": 300941,
|
| 1143 |
+
"input_features": 4096,
|
| 1144 |
+
"sample_rows": 8
|
| 1145 |
+
},
|
| 1146 |
+
"model.language_model.layers.6.linear_attn.out_proj": {
|
| 1147 |
+
"count": 300941,
|
| 1148 |
+
"input_features": 4096,
|
| 1149 |
+
"sample_rows": 8
|
| 1150 |
+
},
|
| 1151 |
+
"model.language_model.layers.6.mlp.down_proj": {
|
| 1152 |
+
"count": 300941,
|
| 1153 |
+
"input_features": 12288,
|
| 1154 |
+
"sample_rows": 8
|
| 1155 |
+
},
|
| 1156 |
+
"model.language_model.layers.6.mlp.gate_proj": {
|
| 1157 |
+
"count": 300941,
|
| 1158 |
+
"input_features": 4096,
|
| 1159 |
+
"sample_rows": 8
|
| 1160 |
+
},
|
| 1161 |
+
"model.language_model.layers.6.mlp.up_proj": {
|
| 1162 |
+
"count": 300941,
|
| 1163 |
+
"input_features": 4096,
|
| 1164 |
+
"sample_rows": 8
|
| 1165 |
+
},
|
| 1166 |
+
"model.language_model.layers.7.mlp.down_proj": {
|
| 1167 |
+
"count": 300941,
|
| 1168 |
+
"input_features": 12288,
|
| 1169 |
+
"sample_rows": 8
|
| 1170 |
+
},
|
| 1171 |
+
"model.language_model.layers.7.mlp.gate_proj": {
|
| 1172 |
+
"count": 300941,
|
| 1173 |
+
"input_features": 4096,
|
| 1174 |
+
"sample_rows": 8
|
| 1175 |
+
},
|
| 1176 |
+
"model.language_model.layers.7.mlp.up_proj": {
|
| 1177 |
+
"count": 300941,
|
| 1178 |
+
"input_features": 4096,
|
| 1179 |
+
"sample_rows": 8
|
| 1180 |
+
},
|
| 1181 |
+
"model.language_model.layers.7.self_attn.k_proj": {
|
| 1182 |
+
"count": 300941,
|
| 1183 |
+
"input_features": 4096,
|
| 1184 |
+
"sample_rows": 8
|
| 1185 |
+
},
|
| 1186 |
+
"model.language_model.layers.7.self_attn.o_proj": {
|
| 1187 |
+
"count": 300941,
|
| 1188 |
+
"input_features": 4096,
|
| 1189 |
+
"sample_rows": 8
|
| 1190 |
+
},
|
| 1191 |
+
"model.language_model.layers.7.self_attn.q_proj": {
|
| 1192 |
+
"count": 300941,
|
| 1193 |
+
"input_features": 4096,
|
| 1194 |
+
"sample_rows": 8
|
| 1195 |
+
},
|
| 1196 |
+
"model.language_model.layers.7.self_attn.v_proj": {
|
| 1197 |
+
"count": 300941,
|
| 1198 |
+
"input_features": 4096,
|
| 1199 |
+
"sample_rows": 8
|
| 1200 |
+
},
|
| 1201 |
+
"model.language_model.layers.8.linear_attn.in_proj_a": {
|
| 1202 |
+
"count": 300941,
|
| 1203 |
+
"input_features": 4096,
|
| 1204 |
+
"sample_rows": 8
|
| 1205 |
+
},
|
| 1206 |
+
"model.language_model.layers.8.linear_attn.in_proj_b": {
|
| 1207 |
+
"count": 300941,
|
| 1208 |
+
"input_features": 4096,
|
| 1209 |
+
"sample_rows": 8
|
| 1210 |
+
},
|
| 1211 |
+
"model.language_model.layers.8.linear_attn.in_proj_qkv": {
|
| 1212 |
+
"count": 300941,
|
| 1213 |
+
"input_features": 4096,
|
| 1214 |
+
"sample_rows": 8
|
| 1215 |
+
},
|
| 1216 |
+
"model.language_model.layers.8.linear_attn.in_proj_z": {
|
| 1217 |
+
"count": 300941,
|
| 1218 |
+
"input_features": 4096,
|
| 1219 |
+
"sample_rows": 8
|
| 1220 |
+
},
|
| 1221 |
+
"model.language_model.layers.8.linear_attn.out_proj": {
|
| 1222 |
+
"count": 300941,
|
| 1223 |
+
"input_features": 4096,
|
| 1224 |
+
"sample_rows": 8
|
| 1225 |
+
},
|
| 1226 |
+
"model.language_model.layers.8.mlp.down_proj": {
|
| 1227 |
+
"count": 300941,
|
| 1228 |
+
"input_features": 12288,
|
| 1229 |
+
"sample_rows": 8
|
| 1230 |
+
},
|
| 1231 |
+
"model.language_model.layers.8.mlp.gate_proj": {
|
| 1232 |
+
"count": 300941,
|
| 1233 |
+
"input_features": 4096,
|
| 1234 |
+
"sample_rows": 8
|
| 1235 |
+
},
|
| 1236 |
+
"model.language_model.layers.8.mlp.up_proj": {
|
| 1237 |
+
"count": 300941,
|
| 1238 |
+
"input_features": 4096,
|
| 1239 |
+
"sample_rows": 8
|
| 1240 |
+
},
|
| 1241 |
+
"model.language_model.layers.9.linear_attn.in_proj_a": {
|
| 1242 |
+
"count": 300941,
|
| 1243 |
+
"input_features": 4096,
|
| 1244 |
+
"sample_rows": 8
|
| 1245 |
+
},
|
| 1246 |
+
"model.language_model.layers.9.linear_attn.in_proj_b": {
|
| 1247 |
+
"count": 300941,
|
| 1248 |
+
"input_features": 4096,
|
| 1249 |
+
"sample_rows": 8
|
| 1250 |
+
},
|
| 1251 |
+
"model.language_model.layers.9.linear_attn.in_proj_qkv": {
|
| 1252 |
+
"count": 300941,
|
| 1253 |
+
"input_features": 4096,
|
| 1254 |
+
"sample_rows": 8
|
| 1255 |
+
},
|
| 1256 |
+
"model.language_model.layers.9.linear_attn.in_proj_z": {
|
| 1257 |
+
"count": 300941,
|
| 1258 |
+
"input_features": 4096,
|
| 1259 |
+
"sample_rows": 8
|
| 1260 |
+
},
|
| 1261 |
+
"model.language_model.layers.9.linear_attn.out_proj": {
|
| 1262 |
+
"count": 300941,
|
| 1263 |
+
"input_features": 4096,
|
| 1264 |
+
"sample_rows": 8
|
| 1265 |
+
},
|
| 1266 |
+
"model.language_model.layers.9.mlp.down_proj": {
|
| 1267 |
+
"count": 300941,
|
| 1268 |
+
"input_features": 12288,
|
| 1269 |
+
"sample_rows": 8
|
| 1270 |
+
},
|
| 1271 |
+
"model.language_model.layers.9.mlp.gate_proj": {
|
| 1272 |
+
"count": 300941,
|
| 1273 |
+
"input_features": 4096,
|
| 1274 |
+
"sample_rows": 8
|
| 1275 |
+
},
|
| 1276 |
+
"model.language_model.layers.9.mlp.up_proj": {
|
| 1277 |
+
"count": 300941,
|
| 1278 |
+
"input_features": 4096,
|
| 1279 |
+
"sample_rows": 8
|
| 1280 |
+
}
|
| 1281 |
+
},
|
| 1282 |
+
"native_fp32_tensors_restored": 48,
|
| 1283 |
+
"packages": {
|
| 1284 |
+
"accelerate": "1.7.0",
|
| 1285 |
+
"numpy": "1.26.4",
|
| 1286 |
+
"safetensors": "0.8.0",
|
| 1287 |
+
"torch": "2.11.0+cpu",
|
| 1288 |
+
"transformers": "5.12.1"
|
| 1289 |
+
},
|
| 1290 |
+
"parameter_elements_by_dtype": {
|
| 1291 |
+
"torch.bfloat16": 8953799424,
|
| 1292 |
+
"torch.float32": 3840
|
| 1293 |
+
},
|
| 1294 |
+
"reference_loader_sha256": "286b970571921b3a4ca3aa80a75a55c4191de404b4c95ad84828349bc504e046",
|
| 1295 |
+
"revision_evidence": {
|
| 1296 |
+
"chat_template.jinja": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 1297 |
+
"config.json": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 1298 |
+
"merges.txt": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 1299 |
+
"model.safetensors-00001-of-00004.safetensors": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 1300 |
+
"model.safetensors-00002-of-00004.safetensors": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 1301 |
+
"model.safetensors-00003-of-00004.safetensors": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 1302 |
+
"model.safetensors-00004-of-00004.safetensors": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 1303 |
+
"model.safetensors.index.json": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 1304 |
+
"tokenizer.json": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 1305 |
+
"tokenizer_config.json": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 1306 |
+
"vocab.json": "c202236235762e1c871ad0ccb60c8ee5ba337b9a"
|
| 1307 |
+
},
|
| 1308 |
+
"samples_sha256": "f5fa7dbda17bc7e8c09e6108eb648b49766e434e548786fe53746567049b67e4",
|
| 1309 |
+
"sampling": "Fixed approximately uniformly spaced token indices over the full deterministic record stream, not a first-record activation slice",
|
| 1310 |
+
"schema_version": 1,
|
| 1311 |
+
"script_sha256": "8e7b3e30bf5e74ed59d90b0b2e4e0a33a4985c7f7504fd7bd75d97a96087c5dc",
|
| 1312 |
+
"statistic": "float64 sum of FP32 squared linear-input activations over all full-record token positions; count is unpadded token rows. Mean diagonal second moment = sumsq / count.",
|
| 1313 |
+
"status": "complete",
|
| 1314 |
+
"text_tensor_count": 427,
|
| 1315 |
+
"tokenizer": {
|
| 1316 |
+
"applied_here": false,
|
| 1317 |
+
"file_sha256": {
|
| 1318 |
+
"chat_template.jinja": "a4aee8afcf2e0711942cf848899be66016f8d14a889ff9ede07bca099c28f715",
|
| 1319 |
+
"merges.txt": "a9d356d7bdf1ef4949e3e748e95b8e10ad9d4e2e838eddc38a0a7b6b94d1db8d",
|
| 1320 |
+
"tokenizer.json": "5f9e4d4901a92b997e463c1f46055088b6cca5ca61a6522d1b9f64c4bb81cb42",
|
| 1321 |
+
"tokenizer_config.json": "316230d6a809701f4db5ea8f8fc862bc3a6f3229c937c174e674ff3ca0a64ac8",
|
| 1322 |
+
"vocab.json": "ce99b4cb2983d118806ce0a8b777a35b093e2000a503ebde25853284c9dfa003"
|
| 1323 |
+
},
|
| 1324 |
+
"model_id": "Qwen/Qwen3.5-9B",
|
| 1325 |
+
"revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 1326 |
+
"vocabulary_order": "Original checkpoint lm_head row order, unchanged"
|
| 1327 |
+
},
|
| 1328 |
+
"total_input_tokens": 300941,
|
| 1329 |
+
"total_records": 614,
|
| 1330 |
+
"use": "TT precision sensitivity ranking; importance = sumsq/count. Not llama.cpp imatrix format; no training or quantized-weight optimization.",
|
| 1331 |
+
"weight_bytes_by_component": {
|
| 1332 |
+
"mtp": 486581248,
|
| 1333 |
+
"text": 17907614208,
|
| 1334 |
+
"vision": 912020960
|
| 1335 |
+
},
|
| 1336 |
+
"weight_integrity_note": "Revision sidecars plus config/index hashes; full multi-GB shard content hashes are not recomputed."
|
| 1337 |
+
}
|
checkpoint/LICENSE
ADDED
|
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
Apache License
|
| 3 |
+
Version 2.0, January 2004
|
| 4 |
+
http://www.apache.org/licenses/
|
| 5 |
+
|
| 6 |
+
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
|
| 7 |
+
|
| 8 |
+
1. Definitions.
|
| 9 |
+
|
| 10 |
+
"License" shall mean the terms and conditions for use, reproduction,
|
| 11 |
+
and distribution as defined by Sections 1 through 9 of this document.
|
| 12 |
+
|
| 13 |
+
"Licensor" shall mean the copyright owner or entity authorized by
|
| 14 |
+
the copyright owner that is granting the License.
|
| 15 |
+
|
| 16 |
+
"Legal Entity" shall mean the union of the acting entity and all
|
| 17 |
+
other entities that control, are controlled by, or are under common
|
| 18 |
+
control with that entity. For the purposes of this definition,
|
| 19 |
+
"control" means (i) the power, direct or indirect, to cause the
|
| 20 |
+
direction or management of such entity, whether by contract or
|
| 21 |
+
otherwise, or (ii) ownership of fifty percent (50%) or more of the
|
| 22 |
+
outstanding shares, or (iii) beneficial ownership of such entity.
|
| 23 |
+
|
| 24 |
+
"You" (or "Your") shall mean an individual or Legal Entity
|
| 25 |
+
exercising permissions granted by this License.
|
| 26 |
+
|
| 27 |
+
"Source" form shall mean the preferred form for making modifications,
|
| 28 |
+
including but not limited to software source code, documentation
|
| 29 |
+
source, and configuration files.
|
| 30 |
+
|
| 31 |
+
"Object" form shall mean any form resulting from mechanical
|
| 32 |
+
transformation or translation of a Source form, including but
|
| 33 |
+
not limited to compiled object code, generated documentation,
|
| 34 |
+
and conversions to other media types.
|
| 35 |
+
|
| 36 |
+
"Work" shall mean the work of authorship, whether in Source or
|
| 37 |
+
Object form, made available under the License, as indicated by a
|
| 38 |
+
copyright notice that is included in or attached to the work
|
| 39 |
+
(an example is provided in the Appendix below).
|
| 40 |
+
|
| 41 |
+
"Derivative Works" shall mean any work, whether in Source or Object
|
| 42 |
+
form, that is based on (or derived from) the Work and for which the
|
| 43 |
+
editorial revisions, annotations, elaborations, or other modifications
|
| 44 |
+
represent, as a whole, an original work of authorship. For the purposes
|
| 45 |
+
of this License, Derivative Works shall not include works that remain
|
| 46 |
+
separable from, or merely link (or bind by name) to the interfaces of,
|
| 47 |
+
the Work and Derivative Works thereof.
|
| 48 |
+
|
| 49 |
+
"Contribution" shall mean any work of authorship, including
|
| 50 |
+
the original version of the Work and any modifications or additions
|
| 51 |
+
to that Work or Derivative Works thereof, that is intentionally
|
| 52 |
+
submitted to Licensor for inclusion in the Work by the copyright owner
|
| 53 |
+
or by an individual or Legal Entity authorized to submit on behalf of
|
| 54 |
+
the copyright owner. For the purposes of this definition, "submitted"
|
| 55 |
+
means any form of electronic, verbal, or written communication sent
|
| 56 |
+
to the Licensor or its representatives, including but not limited to
|
| 57 |
+
communication on electronic mailing lists, source code control systems,
|
| 58 |
+
and issue tracking systems that are managed by, or on behalf of, the
|
| 59 |
+
Licensor for the purpose of discussing and improving the Work, but
|
| 60 |
+
excluding communication that is conspicuously marked or otherwise
|
| 61 |
+
designated in writing by the copyright owner as "Not a Contribution."
|
| 62 |
+
|
| 63 |
+
"Contributor" shall mean Licensor and any individual or Legal Entity
|
| 64 |
+
on behalf of whom a Contribution has been received by Licensor and
|
| 65 |
+
subsequently incorporated within the Work.
|
| 66 |
+
|
| 67 |
+
2. Grant of Copyright License. Subject to the terms and conditions of
|
| 68 |
+
this License, each Contributor hereby grants to You a perpetual,
|
| 69 |
+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
| 70 |
+
copyright license to reproduce, prepare Derivative Works of,
|
| 71 |
+
publicly display, publicly perform, sublicense, and distribute the
|
| 72 |
+
Work and such Derivative Works in Source or Object form.
|
| 73 |
+
|
| 74 |
+
3. Grant of Patent License. Subject to the terms and conditions of
|
| 75 |
+
this License, each Contributor hereby grants to You a perpetual,
|
| 76 |
+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
| 77 |
+
(except as stated in this section) patent license to make, have made,
|
| 78 |
+
use, offer to sell, sell, import, and otherwise transfer the Work,
|
| 79 |
+
where such license applies only to those patent claims licensable
|
| 80 |
+
by such Contributor that are necessarily infringed by their
|
| 81 |
+
Contribution(s) alone or by combination of their Contribution(s)
|
| 82 |
+
with the Work to which such Contribution(s) was submitted. If You
|
| 83 |
+
institute patent litigation against any entity (including a
|
| 84 |
+
cross-claim or counterclaim in a lawsuit) alleging that the Work
|
| 85 |
+
or a Contribution incorporated within the Work constitutes direct
|
| 86 |
+
or contributory patent infringement, then any patent licenses
|
| 87 |
+
granted to You under this License for that Work shall terminate
|
| 88 |
+
as of the date such litigation is filed.
|
| 89 |
+
|
| 90 |
+
4. Redistribution. You may reproduce and distribute copies of the
|
| 91 |
+
Work or Derivative Works thereof in any medium, with or without
|
| 92 |
+
modifications, and in Source or Object form, provided that You
|
| 93 |
+
meet the following conditions:
|
| 94 |
+
|
| 95 |
+
(a) You must give any other recipients of the Work or
|
| 96 |
+
Derivative Works a copy of this License; and
|
| 97 |
+
|
| 98 |
+
(b) You must cause any modified files to carry prominent notices
|
| 99 |
+
stating that You changed the files; and
|
| 100 |
+
|
| 101 |
+
(c) You must retain, in the Source form of any Derivative Works
|
| 102 |
+
that You distribute, all copyright, patent, trademark, and
|
| 103 |
+
attribution notices from the Source form of the Work,
|
| 104 |
+
excluding those notices that do not pertain to any part of
|
| 105 |
+
the Derivative Works; and
|
| 106 |
+
|
| 107 |
+
(d) If the Work includes a "NOTICE" text file as part of its
|
| 108 |
+
distribution, then any Derivative Works that You distribute must
|
| 109 |
+
include a readable copy of the attribution notices contained
|
| 110 |
+
within such NOTICE file, excluding those notices that do not
|
| 111 |
+
pertain to any part of the Derivative Works, in at least one
|
| 112 |
+
of the following places: within a NOTICE text file distributed
|
| 113 |
+
as part of the Derivative Works; within the Source form or
|
| 114 |
+
documentation, if provided along with the Derivative Works; or,
|
| 115 |
+
within a display generated by the Derivative Works, if and
|
| 116 |
+
wherever such third-party notices normally appear. The contents
|
| 117 |
+
of the NOTICE file are for informational purposes only and
|
| 118 |
+
do not modify the License. You may add Your own attribution
|
| 119 |
+
notices within Derivative Works that You distribute, alongside
|
| 120 |
+
or as an addendum to the NOTICE text from the Work, provided
|
| 121 |
+
that such additional attribution notices cannot be construed
|
| 122 |
+
as modifying the License.
|
| 123 |
+
|
| 124 |
+
You may add Your own copyright statement to Your modifications and
|
| 125 |
+
may provide additional or different license terms and conditions
|
| 126 |
+
for use, reproduction, or distribution of Your modifications, or
|
| 127 |
+
for any such Derivative Works as a whole, provided Your use,
|
| 128 |
+
reproduction, and distribution of the Work otherwise complies with
|
| 129 |
+
the conditions stated in this License.
|
| 130 |
+
|
| 131 |
+
5. Submission of Contributions. Unless You explicitly state otherwise,
|
| 132 |
+
any Contribution intentionally submitted for inclusion in the Work
|
| 133 |
+
by You to the Licensor shall be under the terms and conditions of
|
| 134 |
+
this License, without any additional terms or conditions.
|
| 135 |
+
Notwithstanding the above, nothing herein shall supersede or modify
|
| 136 |
+
the terms of any separate license agreement you may have executed
|
| 137 |
+
with Licensor regarding such Contributions.
|
| 138 |
+
|
| 139 |
+
6. Trademarks. This License does not grant permission to use the trade
|
| 140 |
+
names, trademarks, service marks, or product names of the Licensor,
|
| 141 |
+
except as required for reasonable and customary use in describing the
|
| 142 |
+
origin of the Work and reproducing the content of the NOTICE file.
|
| 143 |
+
|
| 144 |
+
7. Disclaimer of Warranty. Unless required by applicable law or
|
| 145 |
+
agreed to in writing, Licensor provides the Work (and each
|
| 146 |
+
Contributor provides its Contributions) on an "AS IS" BASIS,
|
| 147 |
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
|
| 148 |
+
implied, including, without limitation, any warranties or conditions
|
| 149 |
+
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
|
| 150 |
+
PARTICULAR PURPOSE. You are solely responsible for determining the
|
| 151 |
+
appropriateness of using or redistributing the Work and assume any
|
| 152 |
+
risks associated with Your exercise of permissions under this License.
|
| 153 |
+
|
| 154 |
+
8. Limitation of Liability. In no event and under no legal theory,
|
| 155 |
+
whether in tort (including negligence), contract, or otherwise,
|
| 156 |
+
unless required by applicable law (such as deliberate and grossly
|
| 157 |
+
negligent acts) or agreed to in writing, shall any Contributor be
|
| 158 |
+
liable to You for damages, including any direct, indirect, special,
|
| 159 |
+
incidental, or consequential damages of any character arising as a
|
| 160 |
+
result of this License or out of the use or inability to use the
|
| 161 |
+
Work (including but not limited to damages for loss of goodwill,
|
| 162 |
+
work stoppage, computer failure or malfunction, or any and all
|
| 163 |
+
other commercial damages or losses), even if such Contributor
|
| 164 |
+
has been advised of the possibility of such damages.
|
| 165 |
+
|
| 166 |
+
9. Accepting Warranty or Additional Liability. While redistributing
|
| 167 |
+
the Work or Derivative Works thereof, You may choose to offer,
|
| 168 |
+
and charge a fee for, acceptance of support, warranty, indemnity,
|
| 169 |
+
or other liability obligations and/or rights consistent with this
|
| 170 |
+
License. However, in accepting such obligations, You may act only
|
| 171 |
+
on Your own behalf and on Your sole responsibility, not on behalf
|
| 172 |
+
of any other Contributor, and only if You agree to indemnify,
|
| 173 |
+
defend, and hold each Contributor harmless for any liability
|
| 174 |
+
incurred by, or claims asserted against, such Contributor by reason
|
| 175 |
+
of your accepting any such warranty or additional liability.
|
| 176 |
+
|
| 177 |
+
END OF TERMS AND CONDITIONS
|
| 178 |
+
|
| 179 |
+
APPENDIX: How to apply the Apache License to your work.
|
| 180 |
+
|
| 181 |
+
To apply the Apache License to your work, attach the following
|
| 182 |
+
boilerplate notice, with the fields enclosed by brackets "[]"
|
| 183 |
+
replaced with your own identifying information. (Don't include
|
| 184 |
+
the brackets!) The text should be enclosed in the appropriate
|
| 185 |
+
comment syntax for the file format. We also recommend that a
|
| 186 |
+
file or class name and description of purpose be included on the
|
| 187 |
+
same "printed page" as the copyright notice for easier
|
| 188 |
+
identification within third-party archives.
|
| 189 |
+
|
| 190 |
+
Copyright 2026 Alibaba Cloud
|
| 191 |
+
|
| 192 |
+
Licensed under the Apache License, Version 2.0 (the "License");
|
| 193 |
+
you may not use this file except in compliance with the License.
|
| 194 |
+
You may obtain a copy of the License at
|
| 195 |
+
|
| 196 |
+
http://www.apache.org/licenses/LICENSE-2.0
|
| 197 |
+
|
| 198 |
+
Unless required by applicable law or agreed to in writing, software
|
| 199 |
+
distributed under the License is distributed on an "AS IS" BASIS,
|
| 200 |
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
| 201 |
+
See the License for the specific language governing permissions and
|
| 202 |
+
limitations under the License.
|
checkpoint/chat_template.jinja
ADDED
|
@@ -0,0 +1,154 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- set image_count = namespace(value=0) %}
|
| 2 |
+
{%- set video_count = namespace(value=0) %}
|
| 3 |
+
{%- macro render_content(content, do_vision_count, is_system_content=false) %}
|
| 4 |
+
{%- if content is string %}
|
| 5 |
+
{{- content }}
|
| 6 |
+
{%- elif content is iterable and content is not mapping %}
|
| 7 |
+
{%- for item in content %}
|
| 8 |
+
{%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
|
| 9 |
+
{%- if is_system_content %}
|
| 10 |
+
{{- raise_exception('System message cannot contain images.') }}
|
| 11 |
+
{%- endif %}
|
| 12 |
+
{%- if do_vision_count %}
|
| 13 |
+
{%- set image_count.value = image_count.value + 1 %}
|
| 14 |
+
{%- endif %}
|
| 15 |
+
{%- if add_vision_id %}
|
| 16 |
+
{{- 'Picture ' ~ image_count.value ~ ': ' }}
|
| 17 |
+
{%- endif %}
|
| 18 |
+
{{- '<|vision_start|><|image_pad|><|vision_end|>' }}
|
| 19 |
+
{%- elif 'video' in item or item.type == 'video' %}
|
| 20 |
+
{%- if is_system_content %}
|
| 21 |
+
{{- raise_exception('System message cannot contain videos.') }}
|
| 22 |
+
{%- endif %}
|
| 23 |
+
{%- if do_vision_count %}
|
| 24 |
+
{%- set video_count.value = video_count.value + 1 %}
|
| 25 |
+
{%- endif %}
|
| 26 |
+
{%- if add_vision_id %}
|
| 27 |
+
{{- 'Video ' ~ video_count.value ~ ': ' }}
|
| 28 |
+
{%- endif %}
|
| 29 |
+
{{- '<|vision_start|><|video_pad|><|vision_end|>' }}
|
| 30 |
+
{%- elif 'text' in item %}
|
| 31 |
+
{{- item.text }}
|
| 32 |
+
{%- else %}
|
| 33 |
+
{{- raise_exception('Unexpected item type in content.') }}
|
| 34 |
+
{%- endif %}
|
| 35 |
+
{%- endfor %}
|
| 36 |
+
{%- elif content is none or content is undefined %}
|
| 37 |
+
{{- '' }}
|
| 38 |
+
{%- else %}
|
| 39 |
+
{{- raise_exception('Unexpected content type.') }}
|
| 40 |
+
{%- endif %}
|
| 41 |
+
{%- endmacro %}
|
| 42 |
+
{%- if not messages %}
|
| 43 |
+
{{- raise_exception('No messages provided.') }}
|
| 44 |
+
{%- endif %}
|
| 45 |
+
{%- if tools and tools is iterable and tools is not mapping %}
|
| 46 |
+
{{- '<|im_start|>system\n' }}
|
| 47 |
+
{{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
|
| 48 |
+
{%- for tool in tools %}
|
| 49 |
+
{{- "\n" }}
|
| 50 |
+
{{- tool | tojson }}
|
| 51 |
+
{%- endfor %}
|
| 52 |
+
{{- "\n</tools>" }}
|
| 53 |
+
{{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
|
| 54 |
+
{%- if messages[0].role == 'system' %}
|
| 55 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 56 |
+
{%- if content %}
|
| 57 |
+
{{- '\n\n' + content }}
|
| 58 |
+
{%- endif %}
|
| 59 |
+
{%- endif %}
|
| 60 |
+
{{- '<|im_end|>\n' }}
|
| 61 |
+
{%- else %}
|
| 62 |
+
{%- if messages[0].role == 'system' %}
|
| 63 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 64 |
+
{{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
|
| 65 |
+
{%- endif %}
|
| 66 |
+
{%- endif %}
|
| 67 |
+
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
| 68 |
+
{%- for message in messages[::-1] %}
|
| 69 |
+
{%- set index = (messages|length - 1) - loop.index0 %}
|
| 70 |
+
{%- if ns.multi_step_tool and message.role == "user" %}
|
| 71 |
+
{%- set content = render_content(message.content, false)|trim %}
|
| 72 |
+
{%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
|
| 73 |
+
{%- set ns.multi_step_tool = false %}
|
| 74 |
+
{%- set ns.last_query_index = index %}
|
| 75 |
+
{%- endif %}
|
| 76 |
+
{%- endif %}
|
| 77 |
+
{%- endfor %}
|
| 78 |
+
{%- if ns.multi_step_tool %}
|
| 79 |
+
{{- raise_exception('No user query found in messages.') }}
|
| 80 |
+
{%- endif %}
|
| 81 |
+
{%- for message in messages %}
|
| 82 |
+
{%- set content = render_content(message.content, true)|trim %}
|
| 83 |
+
{%- if message.role == "system" %}
|
| 84 |
+
{%- if not loop.first %}
|
| 85 |
+
{{- raise_exception('System message must be at the beginning.') }}
|
| 86 |
+
{%- endif %}
|
| 87 |
+
{%- elif message.role == "user" %}
|
| 88 |
+
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
| 89 |
+
{%- elif message.role == "assistant" %}
|
| 90 |
+
{%- set reasoning_content = '' %}
|
| 91 |
+
{%- if message.reasoning_content is string %}
|
| 92 |
+
{%- set reasoning_content = message.reasoning_content %}
|
| 93 |
+
{%- else %}
|
| 94 |
+
{%- if '</think>' in content %}
|
| 95 |
+
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
| 96 |
+
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
| 97 |
+
{%- endif %}
|
| 98 |
+
{%- endif %}
|
| 99 |
+
{%- set reasoning_content = reasoning_content|trim %}
|
| 100 |
+
{%- if loop.index0 > ns.last_query_index %}
|
| 101 |
+
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
|
| 102 |
+
{%- else %}
|
| 103 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 104 |
+
{%- endif %}
|
| 105 |
+
{%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
|
| 106 |
+
{%- for tool_call in message.tool_calls %}
|
| 107 |
+
{%- if tool_call.function is defined %}
|
| 108 |
+
{%- set tool_call = tool_call.function %}
|
| 109 |
+
{%- endif %}
|
| 110 |
+
{%- if loop.first %}
|
| 111 |
+
{%- if content|trim %}
|
| 112 |
+
{{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 113 |
+
{%- else %}
|
| 114 |
+
{{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 115 |
+
{%- endif %}
|
| 116 |
+
{%- else %}
|
| 117 |
+
{{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 118 |
+
{%- endif %}
|
| 119 |
+
{%- if tool_call.arguments is defined %}
|
| 120 |
+
{%- for args_name, args_value in tool_call.arguments|items %}
|
| 121 |
+
{{- '<parameter=' + args_name + '>\n' }}
|
| 122 |
+
{%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
|
| 123 |
+
{{- args_value }}
|
| 124 |
+
{{- '\n</parameter>\n' }}
|
| 125 |
+
{%- endfor %}
|
| 126 |
+
{%- endif %}
|
| 127 |
+
{{- '</function>\n</tool_call>' }}
|
| 128 |
+
{%- endfor %}
|
| 129 |
+
{%- endif %}
|
| 130 |
+
{{- '<|im_end|>\n' }}
|
| 131 |
+
{%- elif message.role == "tool" %}
|
| 132 |
+
{%- if loop.previtem and loop.previtem.role != "tool" %}
|
| 133 |
+
{{- '<|im_start|>user' }}
|
| 134 |
+
{%- endif %}
|
| 135 |
+
{{- '\n<tool_response>\n' }}
|
| 136 |
+
{{- content }}
|
| 137 |
+
{{- '\n</tool_response>' }}
|
| 138 |
+
{%- if not loop.last and loop.nextitem.role != "tool" %}
|
| 139 |
+
{{- '<|im_end|>\n' }}
|
| 140 |
+
{%- elif loop.last %}
|
| 141 |
+
{{- '<|im_end|>\n' }}
|
| 142 |
+
{%- endif %}
|
| 143 |
+
{%- else %}
|
| 144 |
+
{{- raise_exception('Unexpected message role.') }}
|
| 145 |
+
{%- endif %}
|
| 146 |
+
{%- endfor %}
|
| 147 |
+
{%- if add_generation_prompt %}
|
| 148 |
+
{{- '<|im_start|>assistant\n' }}
|
| 149 |
+
{%- if enable_thinking is defined and enable_thinking is false %}
|
| 150 |
+
{{- '<think>\n\n</think>\n\n' }}
|
| 151 |
+
{%- else %}
|
| 152 |
+
{{- '<think>\n' }}
|
| 153 |
+
{%- endif %}
|
| 154 |
+
{%- endif %}
|
checkpoint/config.json
ADDED
|
@@ -0,0 +1,103 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3_5ForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"image_token_id": 248056,
|
| 6 |
+
"model_type": "qwen3_5",
|
| 7 |
+
"text_config": {
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"attn_output_gate": true,
|
| 11 |
+
"dtype": "bfloat16",
|
| 12 |
+
"eos_token_id": 248044,
|
| 13 |
+
"full_attention_interval": 4,
|
| 14 |
+
"head_dim": 256,
|
| 15 |
+
"hidden_act": "silu",
|
| 16 |
+
"hidden_size": 4096,
|
| 17 |
+
"initializer_range": 0.02,
|
| 18 |
+
"intermediate_size": 12288,
|
| 19 |
+
"layer_types": [
|
| 20 |
+
"linear_attention",
|
| 21 |
+
"linear_attention",
|
| 22 |
+
"linear_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"linear_attention",
|
| 25 |
+
"linear_attention",
|
| 26 |
+
"linear_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"linear_attention",
|
| 29 |
+
"linear_attention",
|
| 30 |
+
"linear_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"linear_attention",
|
| 33 |
+
"linear_attention",
|
| 34 |
+
"linear_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"linear_attention",
|
| 37 |
+
"linear_attention",
|
| 38 |
+
"linear_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"linear_attention",
|
| 41 |
+
"linear_attention",
|
| 42 |
+
"linear_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"linear_attention",
|
| 45 |
+
"linear_attention",
|
| 46 |
+
"linear_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"linear_attention",
|
| 49 |
+
"linear_attention",
|
| 50 |
+
"linear_attention",
|
| 51 |
+
"full_attention"
|
| 52 |
+
],
|
| 53 |
+
"linear_conv_kernel_dim": 4,
|
| 54 |
+
"linear_key_head_dim": 128,
|
| 55 |
+
"linear_num_key_heads": 16,
|
| 56 |
+
"linear_num_value_heads": 32,
|
| 57 |
+
"linear_value_head_dim": 128,
|
| 58 |
+
"max_position_embeddings": 262144,
|
| 59 |
+
"mlp_only_layers": [],
|
| 60 |
+
"model_type": "qwen3_5_text",
|
| 61 |
+
"mtp_num_hidden_layers": 1,
|
| 62 |
+
"mtp_use_dedicated_embeddings": false,
|
| 63 |
+
"num_attention_heads": 16,
|
| 64 |
+
"num_hidden_layers": 32,
|
| 65 |
+
"num_key_value_heads": 4,
|
| 66 |
+
"rms_norm_eps": 1e-06,
|
| 67 |
+
"use_cache": true,
|
| 68 |
+
"vocab_size": 248320,
|
| 69 |
+
"mamba_ssm_dtype": "float32",
|
| 70 |
+
"rope_parameters": {
|
| 71 |
+
"mrope_interleaved": true,
|
| 72 |
+
"mrope_section": [
|
| 73 |
+
11,
|
| 74 |
+
11,
|
| 75 |
+
10
|
| 76 |
+
],
|
| 77 |
+
"rope_type": "default",
|
| 78 |
+
"rope_theta": 10000000,
|
| 79 |
+
"partial_rotary_factor": 0.25
|
| 80 |
+
}
|
| 81 |
+
},
|
| 82 |
+
"tie_word_embeddings": false,
|
| 83 |
+
"transformers_version": "4.57.0.dev0",
|
| 84 |
+
"video_token_id": 248057,
|
| 85 |
+
"vision_config": {
|
| 86 |
+
"deepstack_visual_indexes": [],
|
| 87 |
+
"depth": 27,
|
| 88 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 89 |
+
"hidden_size": 1152,
|
| 90 |
+
"in_channels": 3,
|
| 91 |
+
"initializer_range": 0.02,
|
| 92 |
+
"intermediate_size": 4304,
|
| 93 |
+
"model_type": "qwen3_5",
|
| 94 |
+
"num_heads": 16,
|
| 95 |
+
"num_position_embeddings": 2304,
|
| 96 |
+
"out_hidden_size": 4096,
|
| 97 |
+
"patch_size": 16,
|
| 98 |
+
"spatial_merge_size": 2,
|
| 99 |
+
"temporal_patch_size": 2
|
| 100 |
+
},
|
| 101 |
+
"vision_end_token_id": 248054,
|
| 102 |
+
"vision_start_token_id": 248053
|
| 103 |
+
}
|
checkpoint/equivalence-mtp.json
ADDED
|
@@ -0,0 +1,342 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"baseline_metadata_sha256": "fff7f404107f42f3db20c52eb4398e2016f00fe8878ef86ebfc114a1b147698b",
|
| 3 |
+
"exact_logits_equal": true,
|
| 4 |
+
"manifest_sha256": "34f774cf110e3ff94726171b52f6d462cb9b36651295b941efa85098ddb96ca2",
|
| 5 |
+
"precision": {
|
| 6 |
+
"layers": {
|
| 7 |
+
"0": {
|
| 8 |
+
"gdn": "bfp8",
|
| 9 |
+
"gdn_output": "bfp8",
|
| 10 |
+
"mlp_down": "bfp4",
|
| 11 |
+
"mlp_gate_up": "bfp4"
|
| 12 |
+
},
|
| 13 |
+
"1": {
|
| 14 |
+
"gdn": "bfp8",
|
| 15 |
+
"gdn_output": "bfp8",
|
| 16 |
+
"mlp_down": "bfp8",
|
| 17 |
+
"mlp_gate_up": "bfp4"
|
| 18 |
+
},
|
| 19 |
+
"10": {
|
| 20 |
+
"gdn": "bfp8",
|
| 21 |
+
"gdn_output": "bfp8",
|
| 22 |
+
"mlp_down": "bfp4",
|
| 23 |
+
"mlp_gate_up": "bfp4"
|
| 24 |
+
},
|
| 25 |
+
"11": {
|
| 26 |
+
"attention": "bfp8",
|
| 27 |
+
"mlp_down": "bfp4",
|
| 28 |
+
"mlp_gate_up": "bfp4"
|
| 29 |
+
},
|
| 30 |
+
"12": {
|
| 31 |
+
"gdn": "bfp8",
|
| 32 |
+
"gdn_output": "bfp8",
|
| 33 |
+
"mlp_down": "bfp4",
|
| 34 |
+
"mlp_gate_up": "bfp8"
|
| 35 |
+
},
|
| 36 |
+
"13": {
|
| 37 |
+
"gdn": "bfp8",
|
| 38 |
+
"gdn_output": "bfp8",
|
| 39 |
+
"mlp_down": "bfp4",
|
| 40 |
+
"mlp_gate_up": "bfp8"
|
| 41 |
+
},
|
| 42 |
+
"14": {
|
| 43 |
+
"gdn": "bfp8",
|
| 44 |
+
"gdn_output": "bfp8",
|
| 45 |
+
"mlp_down": "bfp4",
|
| 46 |
+
"mlp_gate_up": "bfp8"
|
| 47 |
+
},
|
| 48 |
+
"15": {
|
| 49 |
+
"attention": "bfp8",
|
| 50 |
+
"mlp_down": "bfp4",
|
| 51 |
+
"mlp_gate_up": "bfp8"
|
| 52 |
+
},
|
| 53 |
+
"16": {
|
| 54 |
+
"gdn": "bfp8",
|
| 55 |
+
"gdn_output": "bfp8",
|
| 56 |
+
"mlp_down": "bfp8",
|
| 57 |
+
"mlp_gate_up": "bfp8"
|
| 58 |
+
},
|
| 59 |
+
"17": {
|
| 60 |
+
"gdn": "bfp8",
|
| 61 |
+
"gdn_output": "bfp8",
|
| 62 |
+
"mlp_down": "bfp4",
|
| 63 |
+
"mlp_gate_up": "bfp8"
|
| 64 |
+
},
|
| 65 |
+
"18": {
|
| 66 |
+
"gdn": "bfp8",
|
| 67 |
+
"gdn_output": "bfp8",
|
| 68 |
+
"mlp_down": "bfp8",
|
| 69 |
+
"mlp_gate_up": "bfp8"
|
| 70 |
+
},
|
| 71 |
+
"19": {
|
| 72 |
+
"attention": "bfp8",
|
| 73 |
+
"mlp_down": "bfp8",
|
| 74 |
+
"mlp_gate_up": "bfp8"
|
| 75 |
+
},
|
| 76 |
+
"2": {
|
| 77 |
+
"gdn": "bfp8",
|
| 78 |
+
"gdn_output": "bfp8",
|
| 79 |
+
"mlp_down": "bfp4",
|
| 80 |
+
"mlp_gate_up": "bfp4"
|
| 81 |
+
},
|
| 82 |
+
"20": {
|
| 83 |
+
"gdn": "bfp8",
|
| 84 |
+
"gdn_output": "bfp8",
|
| 85 |
+
"mlp_down": "bfp4",
|
| 86 |
+
"mlp_gate_up": "bfp8"
|
| 87 |
+
},
|
| 88 |
+
"21": {
|
| 89 |
+
"gdn": "bfp8",
|
| 90 |
+
"gdn_output": "bfp8",
|
| 91 |
+
"mlp_down": "bfp4",
|
| 92 |
+
"mlp_gate_up": "bfp8"
|
| 93 |
+
},
|
| 94 |
+
"22": {
|
| 95 |
+
"gdn": "bfp8",
|
| 96 |
+
"gdn_output": "bfp8",
|
| 97 |
+
"mlp_down": "bfp8",
|
| 98 |
+
"mlp_gate_up": "bfp8"
|
| 99 |
+
},
|
| 100 |
+
"23": {
|
| 101 |
+
"attention": "bfp8",
|
| 102 |
+
"mlp_down": "bfp4",
|
| 103 |
+
"mlp_gate_up": "bfp8"
|
| 104 |
+
},
|
| 105 |
+
"24": {
|
| 106 |
+
"gdn": "bfp8",
|
| 107 |
+
"gdn_output": "bfp8",
|
| 108 |
+
"mlp_down": "bfp4",
|
| 109 |
+
"mlp_gate_up": "bfp4"
|
| 110 |
+
},
|
| 111 |
+
"25": {
|
| 112 |
+
"gdn": "bfp8",
|
| 113 |
+
"gdn_output": "bfp8",
|
| 114 |
+
"mlp_down": "bfp4",
|
| 115 |
+
"mlp_gate_up": "bfp4"
|
| 116 |
+
},
|
| 117 |
+
"26": {
|
| 118 |
+
"gdn": "bfp8",
|
| 119 |
+
"gdn_output": "bfp8",
|
| 120 |
+
"mlp_down": "bfp4",
|
| 121 |
+
"mlp_gate_up": "bfp4"
|
| 122 |
+
},
|
| 123 |
+
"27": {
|
| 124 |
+
"attention": "bfp8",
|
| 125 |
+
"mlp_down": "bfp4",
|
| 126 |
+
"mlp_gate_up": "bfp8"
|
| 127 |
+
},
|
| 128 |
+
"28": {
|
| 129 |
+
"gdn": "bfp8",
|
| 130 |
+
"gdn_output": "bfp8",
|
| 131 |
+
"mlp_down": "bfp4",
|
| 132 |
+
"mlp_gate_up": "bfp8"
|
| 133 |
+
},
|
| 134 |
+
"29": {
|
| 135 |
+
"gdn": "bfp8",
|
| 136 |
+
"gdn_output": "bfp8",
|
| 137 |
+
"mlp_down": "bfp4",
|
| 138 |
+
"mlp_gate_up": "bfp8"
|
| 139 |
+
},
|
| 140 |
+
"3": {
|
| 141 |
+
"attention": "bfp8",
|
| 142 |
+
"mlp_down": "bfp4",
|
| 143 |
+
"mlp_gate_up": "bfp4"
|
| 144 |
+
},
|
| 145 |
+
"30": {
|
| 146 |
+
"gdn": "bfp8",
|
| 147 |
+
"gdn_output": "bfp8",
|
| 148 |
+
"mlp_down": "bfp8",
|
| 149 |
+
"mlp_gate_up": "bfp8"
|
| 150 |
+
},
|
| 151 |
+
"31": {
|
| 152 |
+
"attention": "bfp8",
|
| 153 |
+
"mlp_down": "bfp8",
|
| 154 |
+
"mlp_gate_up": "bfp8"
|
| 155 |
+
},
|
| 156 |
+
"4": {
|
| 157 |
+
"gdn": "bfp8",
|
| 158 |
+
"gdn_output": "bfp8",
|
| 159 |
+
"mlp_down": "bfp4",
|
| 160 |
+
"mlp_gate_up": "bfp4"
|
| 161 |
+
},
|
| 162 |
+
"5": {
|
| 163 |
+
"gdn": "bfp8",
|
| 164 |
+
"gdn_output": "bfp8",
|
| 165 |
+
"mlp_down": "bfp4",
|
| 166 |
+
"mlp_gate_up": "bfp4"
|
| 167 |
+
},
|
| 168 |
+
"6": {
|
| 169 |
+
"gdn": "bfp8",
|
| 170 |
+
"gdn_output": "bfp8",
|
| 171 |
+
"mlp_down": "bfp8",
|
| 172 |
+
"mlp_gate_up": "bfp4"
|
| 173 |
+
},
|
| 174 |
+
"7": {
|
| 175 |
+
"attention": "bfp8",
|
| 176 |
+
"mlp_down": "bfp4",
|
| 177 |
+
"mlp_gate_up": "bfp4"
|
| 178 |
+
},
|
| 179 |
+
"8": {
|
| 180 |
+
"gdn": "bfp8",
|
| 181 |
+
"gdn_output": "bfp8",
|
| 182 |
+
"mlp_down": "bfp4",
|
| 183 |
+
"mlp_gate_up": "bfp4"
|
| 184 |
+
},
|
| 185 |
+
"9": {
|
| 186 |
+
"gdn": "bfp8",
|
| 187 |
+
"gdn_output": "bfp8",
|
| 188 |
+
"mlp_down": "bfp4",
|
| 189 |
+
"mlp_gate_up": "bfp4"
|
| 190 |
+
}
|
| 191 |
+
},
|
| 192 |
+
"lm_head": "bfp8"
|
| 193 |
+
},
|
| 194 |
+
"records": [
|
| 195 |
+
{
|
| 196 |
+
"baseline_sha256": "730b53e084230fe28916b6d25eebfbd65c04c1c23bbc8a76bedd791ca0673146",
|
| 197 |
+
"id": "instruct-code-train-156476",
|
| 198 |
+
"restored_sha256": "730b53e084230fe28916b6d25eebfbd65c04c1c23bbc8a76bedd791ca0673146"
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"baseline_sha256": "d789b431d2bd3cd32da1204889f24005e59922445920141b543a528f6221b34d",
|
| 202 |
+
"id": "instruct-multilingual-test-5",
|
| 203 |
+
"restored_sha256": "d789b431d2bd3cd32da1204889f24005e59922445920141b543a528f6221b34d"
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"baseline_sha256": "a93674e96ea9e4ac6f91d8ccb37935da84e3d179e1c32e80b171a05fb3638e66",
|
| 207 |
+
"id": "instruct-multilingual-test-1160",
|
| 208 |
+
"restored_sha256": "a93674e96ea9e4ac6f91d8ccb37935da84e3d179e1c32e80b171a05fb3638e66"
|
| 209 |
+
}
|
| 210 |
+
],
|
| 211 |
+
"restored_metadata_sha256": "1fba950eac3fab86047d18337e170a0d294982f532d07c025501059c9cd8724d",
|
| 212 |
+
"runtime_environment": {
|
| 213 |
+
"ARCH_NAME": "blackhole",
|
| 214 |
+
"MESH_DEVICE": "P150",
|
| 215 |
+
"QWEN36_BFP4_LM_HEAD": "1",
|
| 216 |
+
"QWEN36_BFP4_MLP_DOWN": "1",
|
| 217 |
+
"QWEN36_BFP4_WEIGHTS": "1",
|
| 218 |
+
"QWEN36_FUSED_FINAL_RESIDUAL_NORM": "1",
|
| 219 |
+
"QWEN36_FUSED_GATE_UP": "1",
|
| 220 |
+
"QWEN36_FUSED_GATE_UP_CORES": "110",
|
| 221 |
+
"QWEN36_FUSED_GATE_UP_PACKED": "1",
|
| 222 |
+
"QWEN36_FUSED_QKV": "1",
|
| 223 |
+
"QWEN36_GATEUP_CB_SLOTS": "2",
|
| 224 |
+
"QWEN36_GATEUP_OUTSTANDING": "1",
|
| 225 |
+
"QWEN36_GATEUP_REQUIRE_PACKED": "1",
|
| 226 |
+
"QWEN36_LM_HEAD_CB_SLOTS": "2",
|
| 227 |
+
"QWEN36_LM_HEAD_OUTSTANDING": "1",
|
| 228 |
+
"QWEN36_LM_HEAD_PACKED": "1",
|
| 229 |
+
"QWEN36_LM_HEAD_REQUIRE_PACKED": "1",
|
| 230 |
+
"QWEN36_MAX_TOKENS_ALL_USERS": "8192",
|
| 231 |
+
"QWEN36_MLP_DOWN_CB_SLOTS": "2",
|
| 232 |
+
"QWEN36_MLP_DOWN_OUTSTANDING": "1",
|
| 233 |
+
"QWEN36_MLP_DOWN_PACKED": "1",
|
| 234 |
+
"QWEN36_MLP_DOWN_REQUIRE_PACKED": "1",
|
| 235 |
+
"QWEN36_MTP": "1",
|
| 236 |
+
"QWEN36_SHARDED_FINAL_NORM": "1",
|
| 237 |
+
"QWEN36_SINGLE_1D_DECODE": "1",
|
| 238 |
+
"QWEN_GDN_FRONTEND_FUSED": "1",
|
| 239 |
+
"QWEN_GDN_MEGA_CB_SLOTS": "2",
|
| 240 |
+
"QWEN_GDN_MEGA_CORES": "110",
|
| 241 |
+
"QWEN_GDN_MEGA_OUTSTANDING": "1",
|
| 242 |
+
"QWEN_GDN_MEGA_PACKED": "1",
|
| 243 |
+
"QWEN_GDN_MEGA_REQUIRE_PACKED": "1",
|
| 244 |
+
"QWEN_GDN_OUTPUT_FUSED": "1",
|
| 245 |
+
"QWEN_GDN_RECURRENT_FUSED": "1",
|
| 246 |
+
"QWEN_SDPA_BF8": "0",
|
| 247 |
+
"TT_QWEN35_TEXT_VER": "qwen36_blackhole"
|
| 248 |
+
},
|
| 249 |
+
"runtime_sources": {
|
| 250 |
+
"__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
|
| 251 |
+
"demo/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
|
| 252 |
+
"demo/benchmark_vision.py": "98e1c2924c7387748bccb4db7ab4e4b7186747a281145cc7dca29576e44d8f6c",
|
| 253 |
+
"demo/text_demo.py": "9e3c1d86391312efb3ef6c8f07636d586322e7d22359cfb8d58a9cb1443fc37d",
|
| 254 |
+
"demo/vision_demo.py": "fe25dd9e81c3fc8b2ef3908b613500fd4de84106d9e2005ab753f5b62821f523",
|
| 255 |
+
"tests/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
|
| 256 |
+
"tests/conftest.py": "facc72bc2d4daaa6063efb448360b48d48e936388df0bd9ad84ec1833b85f6b2",
|
| 257 |
+
"tests/test_attention_tp.py": "faac4db41e15dea443bb9185635e94d5c9e968fbb8e4b95009ce874f88192566",
|
| 258 |
+
"tests/test_decode_bucketing.py": "b4b122dfeeb530d460a7bdb20534621009710d3b52bbfb28d475581107303422",
|
| 259 |
+
"tests/test_factory.py": "ddbbfd6ab73db4bfc9b45ae32bc6de44d5d032e8d41cb8d27f39aa0a8ab01d73",
|
| 260 |
+
"tests/test_gdn_tp.py": "4fc8a93b4c122f24ffa7c15be459011b39ff72e0ffe5a3665453507f2a74b8d8",
|
| 261 |
+
"tests/test_generate_tp.py": "2cc65458dd3acc06de9a8da5a2457e9d1dae639dabf7436b456f632fdf6408bd",
|
| 262 |
+
"tests/test_mlp.py": "0c7a32c00de7d42700e78e63a12b5abe8d11ecc41f36a33d8b078ffd7f788d2e",
|
| 263 |
+
"tests/test_mlp_tp.py": "3402b0165e6dedd386d9b57a23df80811c6f4f261c818f4a5e448b04608a245a",
|
| 264 |
+
"tests/test_model.py": "123d57bfe23fd7d2d2f070b62589b1af617733dfb566695b11ee775523c29080",
|
| 265 |
+
"tests/test_model_tp.py": "d3733c41e470165895dce249ecbf86776427bb5f9348f267988bcfce7a1a568b",
|
| 266 |
+
"tests/test_patch_merger.py": "893db166c3aea884e05a2099529e4842e03f2ae3ce45fdeae9b765b502c053a4",
|
| 267 |
+
"tests/test_prefill.py": "b4d0ac8c72656e8dc771bf2a9c231f8dded674a9e5ff14be82e878b3d85f6072",
|
| 268 |
+
"tests/test_rope_tp.py": "cf0c1fa054bd79f639cfe773b0acf4f6bcf53bf01c77911a886064e85fdd6a47",
|
| 269 |
+
"tests/test_sampling.py": "7444ccb142551cdf4190305039b94d7d4622a050dd905a53da3256e8c985c2b5",
|
| 270 |
+
"tests/test_vision_attention.py": "e04c1814d95959aee63d7a6fdc2032762418f1f1956a31bd0dbfaeff033f27dd",
|
| 271 |
+
"tests/test_vision_block.py": "3065caa4c032cc89140986418d1c9ae4cfe06ad8e91b05de950579b17dcb3e6c",
|
| 272 |
+
"tests/test_weight_mapping.py": "beb49e1a62ab665523329fed2da45481803676acd98b705c7c2f5be420f28710",
|
| 273 |
+
"tests/test_wrapped_model.py": "eddaca444bc1bea1c73aa28299739a910192b305b7bdc686fb09f96b1767ac2a",
|
| 274 |
+
"tests/unit/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
|
| 275 |
+
"tests/unit/conftest.py": "7dcbd7c08346a18f41f7f22c8fd63503468bcf69b1c4b6d60737cb02081c770d",
|
| 276 |
+
"tests/unit/test_attention.py": "bc4ff5c0c429bca4607534329e62cbcd84061d0b1a9a02dc7ee6534db47cb1e7",
|
| 277 |
+
"tests/unit/test_embedding.py": "66607195f94c141b153f198e0310dd8a3c20a50e3b7689e9d31b19b1e18e3558",
|
| 278 |
+
"tests/unit/test_gdn.py": "15323b3051d4c466d1a6021681ea37b0e70806358ec619f69419e2b36426b501",
|
| 279 |
+
"tests/unit/test_layer.py": "d6babb72837bec309e047d4c03e91733a1711dad2c0568fc65fb15c6c1d94385",
|
| 280 |
+
"tests/unit/test_lm_head.py": "327f70e8d720f4ca52327ecbc87ec52e8d5969b24ab1bd593916362c51604994",
|
| 281 |
+
"tests/unit/test_mlp.py": "3947b52bd589f1daa46634f70f0e0f47c2715d7d75c05607e8c6222e049ba49f",
|
| 282 |
+
"tests/unit/test_model.py": "5a17473efbce8b70ddcb323dfe10a36208a1ec5e069c72298f9d5a0423343ba6",
|
| 283 |
+
"tests/unit/test_rms_norm.py": "1196bed05216527986a366bd98985ccf0ec04bb6bd692dd5c30b7c6f4017f38e",
|
| 284 |
+
"tests/unit/test_rope.py": "c22883d945579f7365ef50fbfd942c62329e2a4f6a907187d8a56151b57eef0c",
|
| 285 |
+
"tests/unit/test_substate.py": "101215ebc50ea587700c4f6dd64ad6fbd40101efc81a917991de856fc67e607e",
|
| 286 |
+
"tt/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
|
| 287 |
+
"tt/attention/__init__.py": "86fbe666fb79e0b726789175fca14912c6bfb971295355165e97a3e8eed4a991",
|
| 288 |
+
"tt/attention/config.py": "6bdec51939dd9824cdb0c904cae46de82fb148d6f8f4be7ecf6d727f5c0cc28e",
|
| 289 |
+
"tt/attention/decode.py": "73ccc4511896933cc0b40bac9b43c0eec583e747a10b4af622e6916233a115d5",
|
| 290 |
+
"tt/attention/gated_attention.py": "42b9480eb138c3c5628d55432e19d875d6fc1847be9ca46b271979d2194017eb",
|
| 291 |
+
"tt/attention/prefill.py": "1cc028e66492bf13d6d8e8040026c6ae4c82618bab19f1db8a7681929553ad77",
|
| 292 |
+
"tt/attention/rope_tp.py": "b841d519864e702ef0c9b63306a28df8d57af6ad17462b9e282b6912239f6e70",
|
| 293 |
+
"tt/attention/tp.py": "a31ec742f2382af56bb620812859be7c25e675f1115c12e9ab713bc2d8bf0f3e",
|
| 294 |
+
"tt/attention/weights.py": "bd50f4f9bcd820fb44a82293e5c34842c9bad55701600410137d19a0e8033428",
|
| 295 |
+
"tt/common.py": "4fb0b62e54f4da4df2013bb05a4504cbf35563e6c7d7ef1073740dff328e4289",
|
| 296 |
+
"tt/fused_swiglu.py": "0c0d8c94369c6e4b5d246afb1f8d36ccaff4b9510b093a76bd243c1c0b67bf51",
|
| 297 |
+
"tt/fused_swiglu_kernel.cpp": "e4d8f6233ec1e86fb80fb0032e01e9c5f7c48ade6603fca50ad59db1c6b5b7f9",
|
| 298 |
+
"tt/gateup_layout/gateup_layout_reader.cpp": "d27e043b36083293a4841c5f69109aeb89bdbe01e1c11cc00e118af04e8bb24a",
|
| 299 |
+
"tt/gdn/__init__.py": "cd9dade74044020df60b3158c82c7ced484a8012d150ae87dcadd09fe45c82e8",
|
| 300 |
+
"tt/gdn/config.py": "d36dc3dabace49f020e96d3979ffd7fba415e9d244db7a511e2ea6611b14e306",
|
| 301 |
+
"tt/gdn/decode.py": "56dd072ad942f2f59af543521cb3a3981bdddd906976892a9c2c9e821f366644",
|
| 302 |
+
"tt/gdn/fused_chunk.py": "d3721fd705f5dcf513ad90fe21496fe2e475ffb14683a07533f51aba77762d7a",
|
| 303 |
+
"tt/gdn/fused_commit.py": "2a248f57b59a74e5a4ef04ab1e6386f24de4ddfdbff2de3904f6abbc1d2168c0",
|
| 304 |
+
"tt/gdn/fused_commit_kernel.cpp": "f92493aaefe3ca27f8e0006b0f749ce900ccd63f4a37a261a126cebda619b1b1",
|
| 305 |
+
"tt/gdn/fused_verify_output.cpp": "10f0f96fc1a4f9a8888ff606974f4f5ad0309f119be30fb5490c79e1a4584e5c",
|
| 306 |
+
"tt/gdn/fused_verify_output.py": "e782e7711a02eac5de100bcd2b26b463437d0f6ad05ee3b345e319c4c1dff3ce",
|
| 307 |
+
"tt/gdn/fused_verify_writer.cpp": "f38160d095b1fb384a875f2dbd949d5cd0708f8316d708a9e10d3798b41bfad6",
|
| 308 |
+
"tt/gdn/gated_deltanet.py": "5b3b1efad65cf9ddf6c201fff6ced22dda0f8feb924d564047edbb484ca0f006",
|
| 309 |
+
"tt/gdn/native_frontend.py": "45531e65d90e79ddf8d51b7757060ff8e86aaf1b3e80e0f2740c90f7c363a11a",
|
| 310 |
+
"tt/gdn/native_frontend_kernel.cpp": "0d46fe1db394499da78dda8c6f83f9776692388b55afa2ff3cf615d23f4490b3",
|
| 311 |
+
"tt/gdn/native_output.py": "4961e26b7fb0ecd068ec92871389d5502e244f0002dda8fdbec68da5deaab508",
|
| 312 |
+
"tt/gdn/native_output_kernel.cpp": "16088ec9cd0d2001081f58cf595668e5e565f1db0a3f6a93ec654eba965335bd",
|
| 313 |
+
"tt/gdn/state.py": "34a2323786d33640ad9fb8a9d74f8e8f830a8a46e3946c5029648c5647ed944e",
|
| 314 |
+
"tt/gdn/tp.py": "72c1a2fe5d7b60b6740d42037c70ff2fe93a5e1fbaa3b5e18d629e2f4cd47de9",
|
| 315 |
+
"tt/gdn/weights.py": "a7dbd253ab080dac1885e1a6f1bd7831032c2222c7200932c179941f61112302",
|
| 316 |
+
"tt/generator_interface.py": "5eb6f5c1874e88ae097f6be4a884b38790525fa42f62ca88c3763ef490b3c456",
|
| 317 |
+
"tt/layer.py": "fa986f11ee9ca7b22e4387b5b47ac936192513ac96e0c633df266671fddae4fd",
|
| 318 |
+
"tt/mlp.py": "9a9356f88f6124b74967e1fed34bea5383a83ffbf7206e944a0964e70ba1a11f",
|
| 319 |
+
"tt/model.py": "23d23caef5aac2f34b3ac41f9af9b5cd65ae0a67b7424e3d50d9741fc919e684",
|
| 320 |
+
"tt/model_config.py": "268444cc9edcd069e0fa74141a286035512551c40141e50565726d134cc7903c",
|
| 321 |
+
"tt/mtp.py": "6ccd78a5ffd176e7267d60d6cff960014918df203130614a0cd2b201eeffa4c8",
|
| 322 |
+
"tt/qwen36_vllm.py": "2e3bc11507f4fbba00464593e048faa1e29d8928722143839541c96e319be7fc",
|
| 323 |
+
"tt/rms_norm.py": "5bb650d835e5ed6cdfa375e64c80ed92a1758996883ffb7c238b2195869f6f00",
|
| 324 |
+
"tt/rope.py": "baa056d37129f1fa97a444ba5f6d5f3a5cfd6c15402dcfad13d300b11f2cef26",
|
| 325 |
+
"tt/tp_common.py": "bb43f0cde336c3f84725d47a64ed2b506b5287bdd0e910cd24b13feed0a0826a",
|
| 326 |
+
"tt/vision/__init__.py": "cc2c73418c0211e5db739ce19e47b7c479f507d246e37239d90d0f469d73f1ef",
|
| 327 |
+
"tt/vision/functional.py": "02a48050552879762da15634627500c04e49c620ef880c05c794715c536e5511",
|
| 328 |
+
"tt/vision/model.py": "92f33dcd670d8bf2000a8f523b2dcba4cada9599d309e0d7d079a87720742f91",
|
| 329 |
+
"tt/vision/patch_merger.py": "c7af2a141d45c8d7843445aaa0f68be9eaf5f6249d1c42c2a8f0fa393496aae9",
|
| 330 |
+
"tt/vision/vision_attention.py": "5f99bb9243dce7c37045f82d5917b005bb7aa6ca6dfac788bb69abc44927a0be",
|
| 331 |
+
"tt/vision/vision_block.py": "1d951f0732445ee099112711c8a59b9701a235e35668b87b97cb965faa74ad64",
|
| 332 |
+
"tt/vision/vision_distributed_layernorm.py": "47fd0fcf240b4a1dba399e2f608b050ecc789ffe2c7871d375b94574bda49a23",
|
| 333 |
+
"tt/vision/vision_layernorm.py": "cb2b91f9d6cdf788847871eeb3fd21059dc3934ed274e61e40f5e9fc98e0db60",
|
| 334 |
+
"tt/vision/vision_mlp.py": "6c3c092c33df7cba0483b6ab88903f29a95f2fcecc723d3884e3cbe4effcd0e5",
|
| 335 |
+
"tt/vision/vision_model_config.py": "bf1eb6f5d47243c46c33ee2c3506f6985131a379161d71a02b7bff88bdec5a4d",
|
| 336 |
+
"tt/weight_mapping.py": "d7ad88592fb7de07d87fa0d10f9d1bd0d4dcc92856438fffc394f90e29e650d2",
|
| 337 |
+
"utils/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
|
| 338 |
+
"utils/substate.py": "813aa26dc9053e1217425a3cacf0c63255254f9d17723f3b5ac378167cfca28c"
|
| 339 |
+
},
|
| 340 |
+
"scope": "Exact parity only on recorded full-vocabulary teacher-forced sequences; not a quality certification",
|
| 341 |
+
"tokens_compared": 1415
|
| 342 |
+
}
|
checkpoint/equivalence.json
ADDED
|
@@ -0,0 +1,342 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"baseline_metadata_sha256": "cf8d5c51f40fe78c935b71dfacd821073dca741ed28db78f6794e9cb020870e1",
|
| 3 |
+
"exact_logits_equal": true,
|
| 4 |
+
"manifest_sha256": "34f774cf110e3ff94726171b52f6d462cb9b36651295b941efa85098ddb96ca2",
|
| 5 |
+
"precision": {
|
| 6 |
+
"layers": {
|
| 7 |
+
"0": {
|
| 8 |
+
"gdn": "bfp8",
|
| 9 |
+
"gdn_output": "bfp8",
|
| 10 |
+
"mlp_down": "bfp4",
|
| 11 |
+
"mlp_gate_up": "bfp4"
|
| 12 |
+
},
|
| 13 |
+
"1": {
|
| 14 |
+
"gdn": "bfp8",
|
| 15 |
+
"gdn_output": "bfp8",
|
| 16 |
+
"mlp_down": "bfp8",
|
| 17 |
+
"mlp_gate_up": "bfp4"
|
| 18 |
+
},
|
| 19 |
+
"10": {
|
| 20 |
+
"gdn": "bfp8",
|
| 21 |
+
"gdn_output": "bfp8",
|
| 22 |
+
"mlp_down": "bfp4",
|
| 23 |
+
"mlp_gate_up": "bfp4"
|
| 24 |
+
},
|
| 25 |
+
"11": {
|
| 26 |
+
"attention": "bfp8",
|
| 27 |
+
"mlp_down": "bfp4",
|
| 28 |
+
"mlp_gate_up": "bfp4"
|
| 29 |
+
},
|
| 30 |
+
"12": {
|
| 31 |
+
"gdn": "bfp8",
|
| 32 |
+
"gdn_output": "bfp8",
|
| 33 |
+
"mlp_down": "bfp4",
|
| 34 |
+
"mlp_gate_up": "bfp8"
|
| 35 |
+
},
|
| 36 |
+
"13": {
|
| 37 |
+
"gdn": "bfp8",
|
| 38 |
+
"gdn_output": "bfp8",
|
| 39 |
+
"mlp_down": "bfp4",
|
| 40 |
+
"mlp_gate_up": "bfp8"
|
| 41 |
+
},
|
| 42 |
+
"14": {
|
| 43 |
+
"gdn": "bfp8",
|
| 44 |
+
"gdn_output": "bfp8",
|
| 45 |
+
"mlp_down": "bfp4",
|
| 46 |
+
"mlp_gate_up": "bfp8"
|
| 47 |
+
},
|
| 48 |
+
"15": {
|
| 49 |
+
"attention": "bfp8",
|
| 50 |
+
"mlp_down": "bfp4",
|
| 51 |
+
"mlp_gate_up": "bfp8"
|
| 52 |
+
},
|
| 53 |
+
"16": {
|
| 54 |
+
"gdn": "bfp8",
|
| 55 |
+
"gdn_output": "bfp8",
|
| 56 |
+
"mlp_down": "bfp8",
|
| 57 |
+
"mlp_gate_up": "bfp8"
|
| 58 |
+
},
|
| 59 |
+
"17": {
|
| 60 |
+
"gdn": "bfp8",
|
| 61 |
+
"gdn_output": "bfp8",
|
| 62 |
+
"mlp_down": "bfp4",
|
| 63 |
+
"mlp_gate_up": "bfp8"
|
| 64 |
+
},
|
| 65 |
+
"18": {
|
| 66 |
+
"gdn": "bfp8",
|
| 67 |
+
"gdn_output": "bfp8",
|
| 68 |
+
"mlp_down": "bfp8",
|
| 69 |
+
"mlp_gate_up": "bfp8"
|
| 70 |
+
},
|
| 71 |
+
"19": {
|
| 72 |
+
"attention": "bfp8",
|
| 73 |
+
"mlp_down": "bfp8",
|
| 74 |
+
"mlp_gate_up": "bfp8"
|
| 75 |
+
},
|
| 76 |
+
"2": {
|
| 77 |
+
"gdn": "bfp8",
|
| 78 |
+
"gdn_output": "bfp8",
|
| 79 |
+
"mlp_down": "bfp4",
|
| 80 |
+
"mlp_gate_up": "bfp4"
|
| 81 |
+
},
|
| 82 |
+
"20": {
|
| 83 |
+
"gdn": "bfp8",
|
| 84 |
+
"gdn_output": "bfp8",
|
| 85 |
+
"mlp_down": "bfp4",
|
| 86 |
+
"mlp_gate_up": "bfp8"
|
| 87 |
+
},
|
| 88 |
+
"21": {
|
| 89 |
+
"gdn": "bfp8",
|
| 90 |
+
"gdn_output": "bfp8",
|
| 91 |
+
"mlp_down": "bfp4",
|
| 92 |
+
"mlp_gate_up": "bfp8"
|
| 93 |
+
},
|
| 94 |
+
"22": {
|
| 95 |
+
"gdn": "bfp8",
|
| 96 |
+
"gdn_output": "bfp8",
|
| 97 |
+
"mlp_down": "bfp8",
|
| 98 |
+
"mlp_gate_up": "bfp8"
|
| 99 |
+
},
|
| 100 |
+
"23": {
|
| 101 |
+
"attention": "bfp8",
|
| 102 |
+
"mlp_down": "bfp4",
|
| 103 |
+
"mlp_gate_up": "bfp8"
|
| 104 |
+
},
|
| 105 |
+
"24": {
|
| 106 |
+
"gdn": "bfp8",
|
| 107 |
+
"gdn_output": "bfp8",
|
| 108 |
+
"mlp_down": "bfp4",
|
| 109 |
+
"mlp_gate_up": "bfp4"
|
| 110 |
+
},
|
| 111 |
+
"25": {
|
| 112 |
+
"gdn": "bfp8",
|
| 113 |
+
"gdn_output": "bfp8",
|
| 114 |
+
"mlp_down": "bfp4",
|
| 115 |
+
"mlp_gate_up": "bfp4"
|
| 116 |
+
},
|
| 117 |
+
"26": {
|
| 118 |
+
"gdn": "bfp8",
|
| 119 |
+
"gdn_output": "bfp8",
|
| 120 |
+
"mlp_down": "bfp4",
|
| 121 |
+
"mlp_gate_up": "bfp4"
|
| 122 |
+
},
|
| 123 |
+
"27": {
|
| 124 |
+
"attention": "bfp8",
|
| 125 |
+
"mlp_down": "bfp4",
|
| 126 |
+
"mlp_gate_up": "bfp8"
|
| 127 |
+
},
|
| 128 |
+
"28": {
|
| 129 |
+
"gdn": "bfp8",
|
| 130 |
+
"gdn_output": "bfp8",
|
| 131 |
+
"mlp_down": "bfp4",
|
| 132 |
+
"mlp_gate_up": "bfp8"
|
| 133 |
+
},
|
| 134 |
+
"29": {
|
| 135 |
+
"gdn": "bfp8",
|
| 136 |
+
"gdn_output": "bfp8",
|
| 137 |
+
"mlp_down": "bfp4",
|
| 138 |
+
"mlp_gate_up": "bfp8"
|
| 139 |
+
},
|
| 140 |
+
"3": {
|
| 141 |
+
"attention": "bfp8",
|
| 142 |
+
"mlp_down": "bfp4",
|
| 143 |
+
"mlp_gate_up": "bfp4"
|
| 144 |
+
},
|
| 145 |
+
"30": {
|
| 146 |
+
"gdn": "bfp8",
|
| 147 |
+
"gdn_output": "bfp8",
|
| 148 |
+
"mlp_down": "bfp8",
|
| 149 |
+
"mlp_gate_up": "bfp8"
|
| 150 |
+
},
|
| 151 |
+
"31": {
|
| 152 |
+
"attention": "bfp8",
|
| 153 |
+
"mlp_down": "bfp8",
|
| 154 |
+
"mlp_gate_up": "bfp8"
|
| 155 |
+
},
|
| 156 |
+
"4": {
|
| 157 |
+
"gdn": "bfp8",
|
| 158 |
+
"gdn_output": "bfp8",
|
| 159 |
+
"mlp_down": "bfp4",
|
| 160 |
+
"mlp_gate_up": "bfp4"
|
| 161 |
+
},
|
| 162 |
+
"5": {
|
| 163 |
+
"gdn": "bfp8",
|
| 164 |
+
"gdn_output": "bfp8",
|
| 165 |
+
"mlp_down": "bfp4",
|
| 166 |
+
"mlp_gate_up": "bfp4"
|
| 167 |
+
},
|
| 168 |
+
"6": {
|
| 169 |
+
"gdn": "bfp8",
|
| 170 |
+
"gdn_output": "bfp8",
|
| 171 |
+
"mlp_down": "bfp8",
|
| 172 |
+
"mlp_gate_up": "bfp4"
|
| 173 |
+
},
|
| 174 |
+
"7": {
|
| 175 |
+
"attention": "bfp8",
|
| 176 |
+
"mlp_down": "bfp4",
|
| 177 |
+
"mlp_gate_up": "bfp4"
|
| 178 |
+
},
|
| 179 |
+
"8": {
|
| 180 |
+
"gdn": "bfp8",
|
| 181 |
+
"gdn_output": "bfp8",
|
| 182 |
+
"mlp_down": "bfp4",
|
| 183 |
+
"mlp_gate_up": "bfp4"
|
| 184 |
+
},
|
| 185 |
+
"9": {
|
| 186 |
+
"gdn": "bfp8",
|
| 187 |
+
"gdn_output": "bfp8",
|
| 188 |
+
"mlp_down": "bfp4",
|
| 189 |
+
"mlp_gate_up": "bfp4"
|
| 190 |
+
}
|
| 191 |
+
},
|
| 192 |
+
"lm_head": "bfp8"
|
| 193 |
+
},
|
| 194 |
+
"records": [
|
| 195 |
+
{
|
| 196 |
+
"baseline_sha256": "730b53e084230fe28916b6d25eebfbd65c04c1c23bbc8a76bedd791ca0673146",
|
| 197 |
+
"id": "instruct-code-train-156476",
|
| 198 |
+
"restored_sha256": "730b53e084230fe28916b6d25eebfbd65c04c1c23bbc8a76bedd791ca0673146"
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"baseline_sha256": "d789b431d2bd3cd32da1204889f24005e59922445920141b543a528f6221b34d",
|
| 202 |
+
"id": "instruct-multilingual-test-5",
|
| 203 |
+
"restored_sha256": "d789b431d2bd3cd32da1204889f24005e59922445920141b543a528f6221b34d"
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"baseline_sha256": "a93674e96ea9e4ac6f91d8ccb37935da84e3d179e1c32e80b171a05fb3638e66",
|
| 207 |
+
"id": "instruct-multilingual-test-1160",
|
| 208 |
+
"restored_sha256": "a93674e96ea9e4ac6f91d8ccb37935da84e3d179e1c32e80b171a05fb3638e66"
|
| 209 |
+
}
|
| 210 |
+
],
|
| 211 |
+
"restored_metadata_sha256": "5409a61cd263cd7f5cb9c89953144502024108d9d1b9e4cfc440a71634805c4c",
|
| 212 |
+
"runtime_environment": {
|
| 213 |
+
"ARCH_NAME": "blackhole",
|
| 214 |
+
"MESH_DEVICE": "P150",
|
| 215 |
+
"QWEN36_BFP4_LM_HEAD": "1",
|
| 216 |
+
"QWEN36_BFP4_MLP_DOWN": "1",
|
| 217 |
+
"QWEN36_BFP4_WEIGHTS": "1",
|
| 218 |
+
"QWEN36_FUSED_FINAL_RESIDUAL_NORM": "1",
|
| 219 |
+
"QWEN36_FUSED_GATE_UP": "1",
|
| 220 |
+
"QWEN36_FUSED_GATE_UP_CORES": "110",
|
| 221 |
+
"QWEN36_FUSED_GATE_UP_PACKED": "1",
|
| 222 |
+
"QWEN36_FUSED_QKV": "1",
|
| 223 |
+
"QWEN36_GATEUP_CB_SLOTS": "2",
|
| 224 |
+
"QWEN36_GATEUP_OUTSTANDING": "1",
|
| 225 |
+
"QWEN36_GATEUP_REQUIRE_PACKED": "1",
|
| 226 |
+
"QWEN36_LM_HEAD_CB_SLOTS": "2",
|
| 227 |
+
"QWEN36_LM_HEAD_OUTSTANDING": "1",
|
| 228 |
+
"QWEN36_LM_HEAD_PACKED": "1",
|
| 229 |
+
"QWEN36_LM_HEAD_REQUIRE_PACKED": "1",
|
| 230 |
+
"QWEN36_MAX_TOKENS_ALL_USERS": "8192",
|
| 231 |
+
"QWEN36_MLP_DOWN_CB_SLOTS": "2",
|
| 232 |
+
"QWEN36_MLP_DOWN_OUTSTANDING": "1",
|
| 233 |
+
"QWEN36_MLP_DOWN_PACKED": "1",
|
| 234 |
+
"QWEN36_MLP_DOWN_REQUIRE_PACKED": "1",
|
| 235 |
+
"QWEN36_MTP": "0",
|
| 236 |
+
"QWEN36_SHARDED_FINAL_NORM": "1",
|
| 237 |
+
"QWEN36_SINGLE_1D_DECODE": "1",
|
| 238 |
+
"QWEN_GDN_FRONTEND_FUSED": "1",
|
| 239 |
+
"QWEN_GDN_MEGA_CB_SLOTS": "2",
|
| 240 |
+
"QWEN_GDN_MEGA_CORES": "110",
|
| 241 |
+
"QWEN_GDN_MEGA_OUTSTANDING": "1",
|
| 242 |
+
"QWEN_GDN_MEGA_PACKED": "1",
|
| 243 |
+
"QWEN_GDN_MEGA_REQUIRE_PACKED": "1",
|
| 244 |
+
"QWEN_GDN_OUTPUT_FUSED": "1",
|
| 245 |
+
"QWEN_GDN_RECURRENT_FUSED": "1",
|
| 246 |
+
"QWEN_SDPA_BF8": "0",
|
| 247 |
+
"TT_QWEN35_TEXT_VER": "qwen36_blackhole"
|
| 248 |
+
},
|
| 249 |
+
"runtime_sources": {
|
| 250 |
+
"__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
|
| 251 |
+
"demo/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
|
| 252 |
+
"demo/benchmark_vision.py": "98e1c2924c7387748bccb4db7ab4e4b7186747a281145cc7dca29576e44d8f6c",
|
| 253 |
+
"demo/text_demo.py": "9e3c1d86391312efb3ef6c8f07636d586322e7d22359cfb8d58a9cb1443fc37d",
|
| 254 |
+
"demo/vision_demo.py": "fe25dd9e81c3fc8b2ef3908b613500fd4de84106d9e2005ab753f5b62821f523",
|
| 255 |
+
"tests/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
|
| 256 |
+
"tests/conftest.py": "facc72bc2d4daaa6063efb448360b48d48e936388df0bd9ad84ec1833b85f6b2",
|
| 257 |
+
"tests/test_attention_tp.py": "faac4db41e15dea443bb9185635e94d5c9e968fbb8e4b95009ce874f88192566",
|
| 258 |
+
"tests/test_decode_bucketing.py": "b4b122dfeeb530d460a7bdb20534621009710d3b52bbfb28d475581107303422",
|
| 259 |
+
"tests/test_factory.py": "ddbbfd6ab73db4bfc9b45ae32bc6de44d5d032e8d41cb8d27f39aa0a8ab01d73",
|
| 260 |
+
"tests/test_gdn_tp.py": "4fc8a93b4c122f24ffa7c15be459011b39ff72e0ffe5a3665453507f2a74b8d8",
|
| 261 |
+
"tests/test_generate_tp.py": "2cc65458dd3acc06de9a8da5a2457e9d1dae639dabf7436b456f632fdf6408bd",
|
| 262 |
+
"tests/test_mlp.py": "0c7a32c00de7d42700e78e63a12b5abe8d11ecc41f36a33d8b078ffd7f788d2e",
|
| 263 |
+
"tests/test_mlp_tp.py": "3402b0165e6dedd386d9b57a23df80811c6f4f261c818f4a5e448b04608a245a",
|
| 264 |
+
"tests/test_model.py": "123d57bfe23fd7d2d2f070b62589b1af617733dfb566695b11ee775523c29080",
|
| 265 |
+
"tests/test_model_tp.py": "d3733c41e470165895dce249ecbf86776427bb5f9348f267988bcfce7a1a568b",
|
| 266 |
+
"tests/test_patch_merger.py": "893db166c3aea884e05a2099529e4842e03f2ae3ce45fdeae9b765b502c053a4",
|
| 267 |
+
"tests/test_prefill.py": "b4d0ac8c72656e8dc771bf2a9c231f8dded674a9e5ff14be82e878b3d85f6072",
|
| 268 |
+
"tests/test_rope_tp.py": "cf0c1fa054bd79f639cfe773b0acf4f6bcf53bf01c77911a886064e85fdd6a47",
|
| 269 |
+
"tests/test_sampling.py": "7444ccb142551cdf4190305039b94d7d4622a050dd905a53da3256e8c985c2b5",
|
| 270 |
+
"tests/test_vision_attention.py": "e04c1814d95959aee63d7a6fdc2032762418f1f1956a31bd0dbfaeff033f27dd",
|
| 271 |
+
"tests/test_vision_block.py": "3065caa4c032cc89140986418d1c9ae4cfe06ad8e91b05de950579b17dcb3e6c",
|
| 272 |
+
"tests/test_weight_mapping.py": "beb49e1a62ab665523329fed2da45481803676acd98b705c7c2f5be420f28710",
|
| 273 |
+
"tests/test_wrapped_model.py": "eddaca444bc1bea1c73aa28299739a910192b305b7bdc686fb09f96b1767ac2a",
|
| 274 |
+
"tests/unit/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
|
| 275 |
+
"tests/unit/conftest.py": "7dcbd7c08346a18f41f7f22c8fd63503468bcf69b1c4b6d60737cb02081c770d",
|
| 276 |
+
"tests/unit/test_attention.py": "bc4ff5c0c429bca4607534329e62cbcd84061d0b1a9a02dc7ee6534db47cb1e7",
|
| 277 |
+
"tests/unit/test_embedding.py": "66607195f94c141b153f198e0310dd8a3c20a50e3b7689e9d31b19b1e18e3558",
|
| 278 |
+
"tests/unit/test_gdn.py": "15323b3051d4c466d1a6021681ea37b0e70806358ec619f69419e2b36426b501",
|
| 279 |
+
"tests/unit/test_layer.py": "d6babb72837bec309e047d4c03e91733a1711dad2c0568fc65fb15c6c1d94385",
|
| 280 |
+
"tests/unit/test_lm_head.py": "327f70e8d720f4ca52327ecbc87ec52e8d5969b24ab1bd593916362c51604994",
|
| 281 |
+
"tests/unit/test_mlp.py": "3947b52bd589f1daa46634f70f0e0f47c2715d7d75c05607e8c6222e049ba49f",
|
| 282 |
+
"tests/unit/test_model.py": "5a17473efbce8b70ddcb323dfe10a36208a1ec5e069c72298f9d5a0423343ba6",
|
| 283 |
+
"tests/unit/test_rms_norm.py": "1196bed05216527986a366bd98985ccf0ec04bb6bd692dd5c30b7c6f4017f38e",
|
| 284 |
+
"tests/unit/test_rope.py": "c22883d945579f7365ef50fbfd942c62329e2a4f6a907187d8a56151b57eef0c",
|
| 285 |
+
"tests/unit/test_substate.py": "101215ebc50ea587700c4f6dd64ad6fbd40101efc81a917991de856fc67e607e",
|
| 286 |
+
"tt/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
|
| 287 |
+
"tt/attention/__init__.py": "86fbe666fb79e0b726789175fca14912c6bfb971295355165e97a3e8eed4a991",
|
| 288 |
+
"tt/attention/config.py": "6bdec51939dd9824cdb0c904cae46de82fb148d6f8f4be7ecf6d727f5c0cc28e",
|
| 289 |
+
"tt/attention/decode.py": "73ccc4511896933cc0b40bac9b43c0eec583e747a10b4af622e6916233a115d5",
|
| 290 |
+
"tt/attention/gated_attention.py": "42b9480eb138c3c5628d55432e19d875d6fc1847be9ca46b271979d2194017eb",
|
| 291 |
+
"tt/attention/prefill.py": "1cc028e66492bf13d6d8e8040026c6ae4c82618bab19f1db8a7681929553ad77",
|
| 292 |
+
"tt/attention/rope_tp.py": "b841d519864e702ef0c9b63306a28df8d57af6ad17462b9e282b6912239f6e70",
|
| 293 |
+
"tt/attention/tp.py": "a31ec742f2382af56bb620812859be7c25e675f1115c12e9ab713bc2d8bf0f3e",
|
| 294 |
+
"tt/attention/weights.py": "bd50f4f9bcd820fb44a82293e5c34842c9bad55701600410137d19a0e8033428",
|
| 295 |
+
"tt/common.py": "4fb0b62e54f4da4df2013bb05a4504cbf35563e6c7d7ef1073740dff328e4289",
|
| 296 |
+
"tt/fused_swiglu.py": "0c0d8c94369c6e4b5d246afb1f8d36ccaff4b9510b093a76bd243c1c0b67bf51",
|
| 297 |
+
"tt/fused_swiglu_kernel.cpp": "e4d8f6233ec1e86fb80fb0032e01e9c5f7c48ade6603fca50ad59db1c6b5b7f9",
|
| 298 |
+
"tt/gateup_layout/gateup_layout_reader.cpp": "d27e043b36083293a4841c5f69109aeb89bdbe01e1c11cc00e118af04e8bb24a",
|
| 299 |
+
"tt/gdn/__init__.py": "cd9dade74044020df60b3158c82c7ced484a8012d150ae87dcadd09fe45c82e8",
|
| 300 |
+
"tt/gdn/config.py": "d36dc3dabace49f020e96d3979ffd7fba415e9d244db7a511e2ea6611b14e306",
|
| 301 |
+
"tt/gdn/decode.py": "56dd072ad942f2f59af543521cb3a3981bdddd906976892a9c2c9e821f366644",
|
| 302 |
+
"tt/gdn/fused_chunk.py": "d3721fd705f5dcf513ad90fe21496fe2e475ffb14683a07533f51aba77762d7a",
|
| 303 |
+
"tt/gdn/fused_commit.py": "2a248f57b59a74e5a4ef04ab1e6386f24de4ddfdbff2de3904f6abbc1d2168c0",
|
| 304 |
+
"tt/gdn/fused_commit_kernel.cpp": "f92493aaefe3ca27f8e0006b0f749ce900ccd63f4a37a261a126cebda619b1b1",
|
| 305 |
+
"tt/gdn/fused_verify_output.cpp": "10f0f96fc1a4f9a8888ff606974f4f5ad0309f119be30fb5490c79e1a4584e5c",
|
| 306 |
+
"tt/gdn/fused_verify_output.py": "e782e7711a02eac5de100bcd2b26b463437d0f6ad05ee3b345e319c4c1dff3ce",
|
| 307 |
+
"tt/gdn/fused_verify_writer.cpp": "f38160d095b1fb384a875f2dbd949d5cd0708f8316d708a9e10d3798b41bfad6",
|
| 308 |
+
"tt/gdn/gated_deltanet.py": "5b3b1efad65cf9ddf6c201fff6ced22dda0f8feb924d564047edbb484ca0f006",
|
| 309 |
+
"tt/gdn/native_frontend.py": "45531e65d90e79ddf8d51b7757060ff8e86aaf1b3e80e0f2740c90f7c363a11a",
|
| 310 |
+
"tt/gdn/native_frontend_kernel.cpp": "0d46fe1db394499da78dda8c6f83f9776692388b55afa2ff3cf615d23f4490b3",
|
| 311 |
+
"tt/gdn/native_output.py": "4961e26b7fb0ecd068ec92871389d5502e244f0002dda8fdbec68da5deaab508",
|
| 312 |
+
"tt/gdn/native_output_kernel.cpp": "16088ec9cd0d2001081f58cf595668e5e565f1db0a3f6a93ec654eba965335bd",
|
| 313 |
+
"tt/gdn/state.py": "34a2323786d33640ad9fb8a9d74f8e8f830a8a46e3946c5029648c5647ed944e",
|
| 314 |
+
"tt/gdn/tp.py": "72c1a2fe5d7b60b6740d42037c70ff2fe93a5e1fbaa3b5e18d629e2f4cd47de9",
|
| 315 |
+
"tt/gdn/weights.py": "a7dbd253ab080dac1885e1a6f1bd7831032c2222c7200932c179941f61112302",
|
| 316 |
+
"tt/generator_interface.py": "5eb6f5c1874e88ae097f6be4a884b38790525fa42f62ca88c3763ef490b3c456",
|
| 317 |
+
"tt/layer.py": "fa986f11ee9ca7b22e4387b5b47ac936192513ac96e0c633df266671fddae4fd",
|
| 318 |
+
"tt/mlp.py": "9a9356f88f6124b74967e1fed34bea5383a83ffbf7206e944a0964e70ba1a11f",
|
| 319 |
+
"tt/model.py": "23d23caef5aac2f34b3ac41f9af9b5cd65ae0a67b7424e3d50d9741fc919e684",
|
| 320 |
+
"tt/model_config.py": "268444cc9edcd069e0fa74141a286035512551c40141e50565726d134cc7903c",
|
| 321 |
+
"tt/mtp.py": "6ccd78a5ffd176e7267d60d6cff960014918df203130614a0cd2b201eeffa4c8",
|
| 322 |
+
"tt/qwen36_vllm.py": "2e3bc11507f4fbba00464593e048faa1e29d8928722143839541c96e319be7fc",
|
| 323 |
+
"tt/rms_norm.py": "5bb650d835e5ed6cdfa375e64c80ed92a1758996883ffb7c238b2195869f6f00",
|
| 324 |
+
"tt/rope.py": "baa056d37129f1fa97a444ba5f6d5f3a5cfd6c15402dcfad13d300b11f2cef26",
|
| 325 |
+
"tt/tp_common.py": "bb43f0cde336c3f84725d47a64ed2b506b5287bdd0e910cd24b13feed0a0826a",
|
| 326 |
+
"tt/vision/__init__.py": "cc2c73418c0211e5db739ce19e47b7c479f507d246e37239d90d0f469d73f1ef",
|
| 327 |
+
"tt/vision/functional.py": "02a48050552879762da15634627500c04e49c620ef880c05c794715c536e5511",
|
| 328 |
+
"tt/vision/model.py": "92f33dcd670d8bf2000a8f523b2dcba4cada9599d309e0d7d079a87720742f91",
|
| 329 |
+
"tt/vision/patch_merger.py": "c7af2a141d45c8d7843445aaa0f68be9eaf5f6249d1c42c2a8f0fa393496aae9",
|
| 330 |
+
"tt/vision/vision_attention.py": "5f99bb9243dce7c37045f82d5917b005bb7aa6ca6dfac788bb69abc44927a0be",
|
| 331 |
+
"tt/vision/vision_block.py": "1d951f0732445ee099112711c8a59b9701a235e35668b87b97cb965faa74ad64",
|
| 332 |
+
"tt/vision/vision_distributed_layernorm.py": "47fd0fcf240b4a1dba399e2f608b050ecc789ffe2c7871d375b94574bda49a23",
|
| 333 |
+
"tt/vision/vision_layernorm.py": "cb2b91f9d6cdf788847871eeb3fd21059dc3934ed274e61e40f5e9fc98e0db60",
|
| 334 |
+
"tt/vision/vision_mlp.py": "6c3c092c33df7cba0483b6ab88903f29a95f2fcecc723d3884e3cbe4effcd0e5",
|
| 335 |
+
"tt/vision/vision_model_config.py": "bf1eb6f5d47243c46c33ee2c3506f6985131a379161d71a02b7bff88bdec5a4d",
|
| 336 |
+
"tt/weight_mapping.py": "d7ad88592fb7de07d87fa0d10f9d1bd0d4dcc92856438fffc394f90e29e650d2",
|
| 337 |
+
"utils/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
|
| 338 |
+
"utils/substate.py": "813aa26dc9053e1217425a3cacf0c63255254f9d17723f3b5ac378167cfca28c"
|
| 339 |
+
},
|
| 340 |
+
"scope": "Exact parity only on recorded full-vocabulary teacher-forced sequences; not a quality certification",
|
| 341 |
+
"tokens_compared": 1415
|
| 342 |
+
}
|
checkpoint/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint/native_checkpoint.py
ADDED
|
@@ -0,0 +1,315 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
#!/usr/bin/env python3
|
| 2 |
+
"""TT-native text-only checkpoint contract and lossless host reconstruction.
|
| 3 |
+
|
| 4 |
+
Native matrices are TTNN TILE dumps in [input, output] orientation. Loading
|
| 5 |
+
returns remapped torch [output, input] BF16 tensors; the runtime requantizes
|
| 6 |
+
them. This is NOT direct device-cache loading, GGUF, QAT, or an official
|
| 7 |
+
Unsloth quant. Exact per-matrix idempotence is necessary but insufficient:
|
| 8 |
+
GDN AB/mega tensors concatenate quant-rounded components, and packed/fused
|
| 9 |
+
layouts can regroup block exponents. End-to-end equivalence is required for
|
| 10 |
+
each supported runtime configuration. Single-device generic or dtype-gated
|
| 11 |
+
mixed packed execution with optional one-token MTP can be verified; vision
|
| 12 |
+
and tensor parallelism are excluded. MTP source tensors are stored losslessly.
|
| 13 |
+
|
| 14 |
+
TTNN host conversion may initialize driver metadata: even host-only commands
|
| 15 |
+
need exclusive device ownership in this environment. No original HF checkpoint
|
| 16 |
+
is used by this loader. The installed compatible TT runtime is still required.
|
| 17 |
+
"""
|
| 18 |
+
import argparse
|
| 19 |
+
import hashlib
|
| 20 |
+
import json
|
| 21 |
+
import os
|
| 22 |
+
from pathlib import Path
|
| 23 |
+
|
| 24 |
+
MANIFEST = "native_manifest.json"
|
| 25 |
+
FORMAT = "qwen35-9b-ttnn-native-text-v1"
|
| 26 |
+
DTYPES = {"bf16": "bfloat16", "bfp8": "bfloat8_b", "bfp4": "bfloat4_b"}
|
| 27 |
+
MTP_KEYS = frozenset({
|
| 28 |
+
"mtp.fc.weight",
|
| 29 |
+
"mtp.norm.weight",
|
| 30 |
+
"mtp.pre_fc_norm_embedding.weight",
|
| 31 |
+
"mtp.pre_fc_norm_hidden.weight",
|
| 32 |
+
"mtp.layers.0.input_layernorm.weight",
|
| 33 |
+
"mtp.layers.0.post_attention_layernorm.weight",
|
| 34 |
+
"mtp.layers.0.self_attn.q_norm.weight",
|
| 35 |
+
"mtp.layers.0.self_attn.k_norm.weight",
|
| 36 |
+
"mtp.layers.0.self_attn.q_proj.weight",
|
| 37 |
+
"mtp.layers.0.self_attn.k_proj.weight",
|
| 38 |
+
"mtp.layers.0.self_attn.v_proj.weight",
|
| 39 |
+
"mtp.layers.0.self_attn.o_proj.weight",
|
| 40 |
+
"mtp.layers.0.mlp.gate_proj.weight",
|
| 41 |
+
"mtp.layers.0.mlp.up_proj.weight",
|
| 42 |
+
"mtp.layers.0.mlp.down_proj.weight",
|
| 43 |
+
})
|
| 44 |
+
|
| 45 |
+
|
| 46 |
+
def tensor_hash(tensor):
|
| 47 |
+
import torch
|
| 48 |
+
|
| 49 |
+
raw = tensor.contiguous().view(torch.uint8).reshape(-1).numpy()
|
| 50 |
+
return hashlib.sha256(memoryview(raw)).hexdigest()
|
| 51 |
+
|
| 52 |
+
|
| 53 |
+
def validate_mtp_keys(keys):
|
| 54 |
+
actual = {name for name in keys if name.startswith("mtp.")}
|
| 55 |
+
if actual != MTP_KEYS:
|
| 56 |
+
raise ValueError(f"Incomplete/unsupported one-layer MTP subtree: missing={sorted(MTP_KEYS - actual)}, unexpected={sorted(actual - MTP_KEYS)}")
|
| 57 |
+
|
| 58 |
+
|
| 59 |
+
def equivalence_filename(environment):
|
| 60 |
+
mode = environment.get("QWEN36_MTP")
|
| 61 |
+
if mode not in ("0", "1"):
|
| 62 |
+
raise ValueError("Equivalence requires explicit QWEN36_MTP=0 or QWEN36_MTP=1")
|
| 63 |
+
return "equivalence-mtp.json" if mode == "1" else "equivalence.json"
|
| 64 |
+
|
| 65 |
+
|
| 66 |
+
def digest(path):
|
| 67 |
+
value = hashlib.sha256()
|
| 68 |
+
with Path(path).open("rb") as stream:
|
| 69 |
+
for block in iter(lambda: stream.read(8 * 1024 * 1024), b""):
|
| 70 |
+
value.update(block)
|
| 71 |
+
return value.hexdigest()
|
| 72 |
+
|
| 73 |
+
|
| 74 |
+
def save_json(path, value):
|
| 75 |
+
path = Path(path)
|
| 76 |
+
temporary = path.with_suffix(path.suffix + ".partial")
|
| 77 |
+
temporary.write_text(json.dumps(value, sort_keys=True, indent=2, allow_nan=False) + "\n")
|
| 78 |
+
temporary.replace(path)
|
| 79 |
+
|
| 80 |
+
|
| 81 |
+
def local_file(root, name):
|
| 82 |
+
root = Path(root).resolve()
|
| 83 |
+
path = (root / name).resolve()
|
| 84 |
+
if not path.is_relative_to(root) or path == root:
|
| 85 |
+
raise ValueError(f"Artifact path escapes checkpoint: {name}")
|
| 86 |
+
return path
|
| 87 |
+
|
| 88 |
+
|
| 89 |
+
def read_manifest(root):
|
| 90 |
+
manifest = json.loads((Path(root) / MANIFEST).read_text())
|
| 91 |
+
if manifest.get("format") != FORMAT or manifest.get("scope") != "text-only-no-vision-with-mtp":
|
| 92 |
+
raise ValueError("Unsupported native checkpoint format/scope")
|
| 93 |
+
if not manifest.get("tensors") or not manifest.get("roundtrip", {}).get("all_passed"):
|
| 94 |
+
raise ValueError("Incomplete native checkpoint or failed tensor roundtrip")
|
| 95 |
+
mtp = manifest.get("mtp", {})
|
| 96 |
+
if (mtp.get("enabled") is not True or type(mtp.get("num_speculative_tokens")) is not int
|
| 97 |
+
or mtp["num_speculative_tokens"] != 1 or type(mtp.get("source_tensor_count")) is not int
|
| 98 |
+
or mtp["source_tensor_count"] != len(MTP_KEYS) or mtp.get("storage") != "lossless"):
|
| 99 |
+
raise ValueError("Checkpoint requires complete lossless MTP-1 metadata")
|
| 100 |
+
validate_mtp_keys(manifest["tensors"])
|
| 101 |
+
for name in MTP_KEYS:
|
| 102 |
+
entry = manifest["tensors"][name]
|
| 103 |
+
source_hash = entry.get("source_tensor_sha256")
|
| 104 |
+
file_info = manifest.get("files", {}).get(entry.get("file"), {})
|
| 105 |
+
if (entry.get("storage") != "safetensors-lossless" or entry.get("source_name") != name
|
| 106 |
+
or not entry.get("source_shape") or entry["source_shape"] != entry.get("shape")
|
| 107 |
+
or not entry.get("source_torch_dtype")
|
| 108 |
+
or entry["source_torch_dtype"] != entry.get("restored_torch_dtype")
|
| 109 |
+
or not isinstance(source_hash, str) or len(source_hash) != 64
|
| 110 |
+
or any(c not in "0123456789abcdef" for c in source_hash)
|
| 111 |
+
or source_hash != entry.get("tensor_sha256")
|
| 112 |
+
or not file_info.get("sha256") or entry.get("sha256") != file_info["sha256"]
|
| 113 |
+
or entry.get("roundtrip", {}).get("exact_values") is not True
|
| 114 |
+
or entry.get("roundtrip", {}).get("exact_dtype") is not True):
|
| 115 |
+
raise ValueError(f"MTP tensor is not bound to lossless original values/dtype: {name}")
|
| 116 |
+
return manifest
|
| 117 |
+
|
| 118 |
+
|
| 119 |
+
def verify_files(root, manifest):
|
| 120 |
+
for name, info in manifest["files"].items():
|
| 121 |
+
path = local_file(root, name)
|
| 122 |
+
if path.stat().st_size != info["bytes"] or digest(path) != info["sha256"]:
|
| 123 |
+
raise ValueError(f"Checkpoint file hash/size mismatch: {name}")
|
| 124 |
+
|
| 125 |
+
|
| 126 |
+
def require_equivalence(root, manifest):
|
| 127 |
+
proof = json.loads((Path(root) / equivalence_filename(os.environ)).read_text())
|
| 128 |
+
if (proof.get("manifest_sha256") != digest(Path(root) / MANIFEST)
|
| 129 |
+
or proof.get("exact_logits_equal") is not True or proof.get("tokens_compared", 0) < 1):
|
| 130 |
+
raise ValueError("Native checkpoint lacks matching exact end-to-end equivalence evidence")
|
| 131 |
+
if proof.get("precision") != manifest["precision"]:
|
| 132 |
+
raise ValueError("Equivalence evidence precision mismatch")
|
| 133 |
+
environment = proof.get("runtime_environment", {})
|
| 134 |
+
if equivalence_filename(environment) != equivalence_filename(os.environ):
|
| 135 |
+
raise ValueError("Equivalence evidence MTP mode mismatch")
|
| 136 |
+
return proof
|
| 137 |
+
|
| 138 |
+
|
| 139 |
+
def validate_runtime(root, runtime_root, num_devices):
|
| 140 |
+
"""Reject using equivalence evidence under a different execution contract."""
|
| 141 |
+
if num_devices != 1:
|
| 142 |
+
raise ValueError("Native checkpoint equivalence covers one device only")
|
| 143 |
+
proof = require_equivalence(root, read_manifest(root))
|
| 144 |
+
for name, value in proof["runtime_environment"].items():
|
| 145 |
+
if os.environ.get(name) != value:
|
| 146 |
+
raise ValueError(f"Runtime environment differs from native equivalence: {name}")
|
| 147 |
+
for name, value in proof["runtime_sources"].items():
|
| 148 |
+
if digest(local_file(runtime_root, name)) != value:
|
| 149 |
+
raise ValueError(f"Runtime source differs from native equivalence: {name}")
|
| 150 |
+
|
| 151 |
+
|
| 152 |
+
def matrix_family(name):
|
| 153 |
+
if name == "output.weight":
|
| 154 |
+
return "lm_head"
|
| 155 |
+
if not name.startswith("layers.") or not name.endswith(".weight"):
|
| 156 |
+
return None
|
| 157 |
+
suffix = ".".join(name.split(".")[2:])
|
| 158 |
+
if suffix in {f"self_attn.{p}_proj.weight" for p in ("q", "k", "v", "o")}:
|
| 159 |
+
return "attention"
|
| 160 |
+
if suffix == "linear_attn.out_proj.weight":
|
| 161 |
+
return "gdn_output"
|
| 162 |
+
if suffix in {f"linear_attn.{p}.weight" for p in ("qkv_proj", "in_proj_a", "in_proj_b", "in_proj_z")}:
|
| 163 |
+
return "gdn"
|
| 164 |
+
if suffix in {"mlp.gate_proj.weight", "mlp.up_proj.weight"}:
|
| 165 |
+
return "mlp_gate_up"
|
| 166 |
+
if suffix == "mlp.down_proj.weight":
|
| 167 |
+
return "mlp_down"
|
| 168 |
+
return None
|
| 169 |
+
|
| 170 |
+
|
| 171 |
+
def tensor_precision(name, plan):
|
| 172 |
+
family = matrix_family(name)
|
| 173 |
+
if family is None:
|
| 174 |
+
return None
|
| 175 |
+
if family == "lm_head":
|
| 176 |
+
return plan[family]
|
| 177 |
+
values = plan["layers"][name.split(".")[1]]
|
| 178 |
+
return values.get("gdn_output", values["gdn"]) if family == "gdn_output" else values[family]
|
| 179 |
+
|
| 180 |
+
|
| 181 |
+
def load_native_checkpoint(root, verify_hashes=True, allow_unverified=False):
|
| 182 |
+
"""Reconstruct text and MTP tensors; allow_unverified is for equivalence runs only.
|
| 183 |
+
|
| 184 |
+
Small/nonlinear tensors (including original FP32 A_log/dt_bias) retain
|
| 185 |
+
original values/dtypes. Embeddings and all MTP tensors are stored losslessly. This loader
|
| 186 |
+
materializes one remapped model, never an HF model or an original copy.
|
| 187 |
+
"""
|
| 188 |
+
import torch
|
| 189 |
+
import ttnn
|
| 190 |
+
from safetensors import safe_open
|
| 191 |
+
|
| 192 |
+
root = Path(root).resolve()
|
| 193 |
+
manifest = read_manifest(root)
|
| 194 |
+
if verify_hashes:
|
| 195 |
+
verify_files(root, manifest)
|
| 196 |
+
if not allow_unverified:
|
| 197 |
+
require_equivalence(root, manifest)
|
| 198 |
+
result = {}
|
| 199 |
+
for name, entry in manifest["tensors"].items():
|
| 200 |
+
path = local_file(root, entry["file"])
|
| 201 |
+
if entry["storage"] == "ttnn-tile":
|
| 202 |
+
if entry["precision"] != tensor_precision(name, manifest["precision"]):
|
| 203 |
+
raise ValueError(f"Tensor precision disagrees with checkpoint plan: {name}")
|
| 204 |
+
tensor = ttnn.load_tensor(str(path))
|
| 205 |
+
if list(tensor.shape) != entry["native_shape"] or tensor.dtype != getattr(ttnn, DTYPES[entry["precision"]]):
|
| 206 |
+
raise ValueError(f"Native tensor shape/dtype mismatch: {name}")
|
| 207 |
+
if tensor.layout != ttnn.TILE_LAYOUT:
|
| 208 |
+
raise ValueError(f"Native tensor layout mismatch: {name}")
|
| 209 |
+
value = ttnn.to_torch(tensor).to(torch.bfloat16).T.contiguous()
|
| 210 |
+
del tensor
|
| 211 |
+
elif entry["storage"] == "safetensors-lossless":
|
| 212 |
+
with safe_open(str(path), framework="pt", device="cpu") as source:
|
| 213 |
+
value = source.get_tensor(name)
|
| 214 |
+
else:
|
| 215 |
+
raise ValueError(f"Unsupported tensor storage: {name}")
|
| 216 |
+
if list(value.shape) != entry["shape"] or str(value.dtype) != entry["restored_torch_dtype"]:
|
| 217 |
+
raise ValueError(f"Restored tensor shape/dtype mismatch: {name}")
|
| 218 |
+
if name in MTP_KEYS and tensor_hash(value) != entry["source_tensor_sha256"]:
|
| 219 |
+
raise ValueError(f"Restored MTP tensor differs from original source hash: {name}")
|
| 220 |
+
result[name] = value
|
| 221 |
+
if not {"tok_embeddings.weight", "output.weight", "norm.weight"} <= result.keys():
|
| 222 |
+
raise ValueError("Native checkpoint is missing required top-level tensors")
|
| 223 |
+
return result
|
| 224 |
+
|
| 225 |
+
|
| 226 |
+
def record_equivalence(root, baseline, restored):
|
| 227 |
+
"""Compare actual full-vocabulary runner artifacts, not weight-error proxies.
|
| 228 |
+
|
| 229 |
+
Baseline must evaluate ORIGINAL weights at the SAME chosen precision and
|
| 230 |
+
runtime, not baseline-bf16 against a mixed candidate. Exact logits equality
|
| 231 |
+
is deliberately strict. Evidence covers the supplied records only.
|
| 232 |
+
The destination is selected by the recorded MTP runtime environment, never
|
| 233 |
+
by the environment of this proof-writing process. MTP proof does not replace
|
| 234 |
+
a separate live speculative-cycle verification.
|
| 235 |
+
"""
|
| 236 |
+
import numpy as np
|
| 237 |
+
|
| 238 |
+
root, baseline, restored = map(Path, (root, baseline, restored))
|
| 239 |
+
manifest = read_manifest(root)
|
| 240 |
+
verify_files(root, manifest)
|
| 241 |
+
left = json.loads((baseline / "metadata.json").read_text())
|
| 242 |
+
right = json.loads((restored / "metadata.json").read_text())
|
| 243 |
+
for metadata in (left, right):
|
| 244 |
+
if metadata.get("status") != "complete" or metadata.get("backend") != "ttnn-native":
|
| 245 |
+
raise ValueError("Both evaluations must be completed TTNN runs")
|
| 246 |
+
if metadata.get("precision") != manifest["precision"] or not metadata.get("full_vocabulary"):
|
| 247 |
+
raise ValueError("Both evaluations must use the exported precision and full logits")
|
| 248 |
+
for key in ("precision", "selection", "input_sha256", "runtime_environment", "alignment", "max_seq_len"):
|
| 249 |
+
if left.get(key) != right.get(key):
|
| 250 |
+
raise ValueError(f"End-to-end comparison configuration mismatch: {key}")
|
| 251 |
+
proof_filename = equivalence_filename(left.get("runtime_environment", {}))
|
| 252 |
+
if left["source_identity"]["sources"] != right["source_identity"]["sources"]:
|
| 253 |
+
raise ValueError("Runtime source hashes differ between equivalence runs")
|
| 254 |
+
if right["source_identity"].get("native_manifest_sha256") != digest(root / MANIFEST):
|
| 255 |
+
raise ValueError("Restored evaluation is not bound to this native manifest")
|
| 256 |
+
if left["source_identity"].get("native_manifest_sha256") is not None:
|
| 257 |
+
raise ValueError("Baseline must use the original HF checkpoint")
|
| 258 |
+
if left["source_identity"].get("index_sha256") != manifest["source"]["original_index_sha256"]:
|
| 259 |
+
raise ValueError("Baseline original checkpoint index differs from exported source")
|
| 260 |
+
for metadata in (left, right):
|
| 261 |
+
if metadata["source_identity"].get("config_sha256") != manifest["files"]["config.json"]["sha256"]:
|
| 262 |
+
raise ValueError("Evaluation config differs from native checkpoint")
|
| 263 |
+
rows_a = [json.loads(line) for line in (baseline / "records.jsonl").read_text().splitlines() if line.strip()]
|
| 264 |
+
rows_b = [json.loads(line) for line in (restored / "records.jsonl").read_text().splitlines() if line.strip()]
|
| 265 |
+
if not rows_a or len(rows_a) != len(rows_b):
|
| 266 |
+
raise ValueError("Evaluation record coverage differs or is empty")
|
| 267 |
+
for metadata, rows in ((left, rows_a), (right, rows_b)):
|
| 268 |
+
if metadata.get("completed_records") != len(rows) or [row["id"] for row in rows] != metadata["selection"]["record_ids"]:
|
| 269 |
+
raise ValueError("Evaluation records do not cover the declared completed selection")
|
| 270 |
+
evidence, tokens = [], 0
|
| 271 |
+
for a, b in zip(rows_a, rows_b):
|
| 272 |
+
for key in ("id", "split", "token_ids"):
|
| 273 |
+
if a[key] != b[key]:
|
| 274 |
+
raise ValueError(f"Evaluation record mismatch: {key}")
|
| 275 |
+
file_a = local_file(baseline, a["logits_file"])
|
| 276 |
+
file_b = local_file(restored, b["logits_file"])
|
| 277 |
+
x, y = np.load(file_a, mmap_mode="r"), np.load(file_b, mmap_mode="r")
|
| 278 |
+
expected = (len(a["token_ids"]) - 1, left["vocab_size"])
|
| 279 |
+
if x.shape != expected or y.shape != expected:
|
| 280 |
+
raise ValueError(f"Unexpected full-logits shape for record {a['id']}")
|
| 281 |
+
for start in range(0, x.shape[0], 16):
|
| 282 |
+
if not np.isfinite(x[start:start + 16]).all() or not np.array_equal(x[start:start + 16], y[start:start + 16]):
|
| 283 |
+
raise ValueError(f"Native reload logits differ for record {a['id']} at chunk {start}")
|
| 284 |
+
tokens += x.shape[0]
|
| 285 |
+
evidence.append({"id": a["id"], "baseline_sha256": digest(file_a), "restored_sha256": digest(file_b)})
|
| 286 |
+
proof = {"manifest_sha256": digest(root / MANIFEST), "exact_logits_equal": True,
|
| 287 |
+
"tokens_compared": tokens, "records": evidence, "precision": manifest["precision"],
|
| 288 |
+
"runtime_sources": left["source_identity"]["sources"],
|
| 289 |
+
"runtime_environment": left["runtime_environment"],
|
| 290 |
+
"baseline_metadata_sha256": digest(baseline / "metadata.json"),
|
| 291 |
+
"restored_metadata_sha256": digest(restored / "metadata.json"),
|
| 292 |
+
"scope": "Exact parity only on recorded full-vocabulary teacher-forced sequences; not a quality certification"}
|
| 293 |
+
save_json(root / proof_filename, proof)
|
| 294 |
+
return proof
|
| 295 |
+
|
| 296 |
+
|
| 297 |
+
def main():
|
| 298 |
+
parser = argparse.ArgumentParser(description=__doc__)
|
| 299 |
+
parser.add_argument("checkpoint", type=Path)
|
| 300 |
+
parser.add_argument("--baseline", type=Path)
|
| 301 |
+
parser.add_argument("--restored", type=Path)
|
| 302 |
+
args = parser.parse_args()
|
| 303 |
+
if bool(args.baseline) != bool(args.restored):
|
| 304 |
+
parser.error("--baseline and --restored must be supplied together")
|
| 305 |
+
if args.baseline:
|
| 306 |
+
print(json.dumps(record_equivalence(args.checkpoint, args.baseline, args.restored), indent=2))
|
| 307 |
+
else:
|
| 308 |
+
manifest = read_manifest(args.checkpoint)
|
| 309 |
+
verify_files(args.checkpoint, manifest)
|
| 310 |
+
print(json.dumps({"files_verified": True, "tensors": len(manifest["tensors"]),
|
| 311 |
+
"equivalence": require_equivalence(args.checkpoint, manifest)}, indent=2))
|
| 312 |
+
|
| 313 |
+
|
| 314 |
+
if __name__ == "__main__":
|
| 315 |
+
main()
|
checkpoint/native_manifest.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint/preprocessor_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"size": {
|
| 3 |
+
"longest_edge": 16777216,
|
| 4 |
+
"shortest_edge": 65536
|
| 5 |
+
},
|
| 6 |
+
"patch_size": 16,
|
| 7 |
+
"temporal_patch_size": 2,
|
| 8 |
+
"merge_size": 2,
|
| 9 |
+
"image_mean": [
|
| 10 |
+
0.5,
|
| 11 |
+
0.5,
|
| 12 |
+
0.5
|
| 13 |
+
],
|
| 14 |
+
"image_std": [
|
| 15 |
+
0.5,
|
| 16 |
+
0.5,
|
| 17 |
+
0.5
|
| 18 |
+
],
|
| 19 |
+
"processor_class": "Qwen3VLProcessor",
|
| 20 |
+
"image_processor_type": "Qwen2VLImageProcessorFast"
|
| 21 |
+
}
|
checkpoint/provenance/build_native_checkpoint.py
ADDED
|
@@ -0,0 +1,277 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
#!/usr/bin/env python3
|
| 2 |
+
"""Stream original Qwen3.5-9B text and MTP safetensors into a standalone TT-native PTQ.
|
| 3 |
+
|
| 4 |
+
Run only with exclusive TT device ownership. TTNN host conversion
|
| 5 |
+
initializes device metadata. Example inside the pinned container:
|
| 6 |
+
python /work/build_native_checkpoint.py --weights /weights/Qwen3.5-9B \
|
| 7 |
+
--output /work/native-candidate --profile current-bfp4 \
|
| 8 |
+
--overrides /work/candidate.json --importance /work/importance/importance.npz \
|
| 9 |
+
--container-image IMAGE_ID --device-ownership-confirmed
|
| 10 |
+
|
| 11 |
+
Each target tensor is remapped individually; linear matrices are transposed
|
| 12 |
+
BEFORE native TILE quantization. Other tensors, including BF16 embeddings and
|
| 13 |
+
original FP32 nonlinear parameters, are preserved as individual safetensors.
|
| 14 |
+
All original mtp.* tensors bypass remapping and quantization, preserving their
|
| 15 |
+
runtime keys, values and source dtype for the existing one-layer MTP runtime.
|
| 16 |
+
Importance only measures/ranks precision error: it does not optimize rounding,
|
| 17 |
+
implement an imatrix-aware quantizer, or measure output KL. Candidate quality
|
| 18 |
+
must be evaluated on held-out text with the separate TT runner/scorer.
|
| 19 |
+
"""
|
| 20 |
+
import argparse
|
| 21 |
+
import datetime
|
| 22 |
+
import importlib.util
|
| 23 |
+
import json
|
| 24 |
+
import platform
|
| 25 |
+
import shutil
|
| 26 |
+
from pathlib import Path
|
| 27 |
+
|
| 28 |
+
from native_checkpoint import DTYPES, FORMAT, MANIFEST, MTP_KEYS, digest, local_file, matrix_family, save_json, tensor_hash, tensor_precision, validate_mtp_keys
|
| 29 |
+
|
| 30 |
+
ROOT = Path(__file__).resolve().parent
|
| 31 |
+
ASSETS = ("config.json", "tokenizer.json", "tokenizer_config.json", "vocab.json", "merges.txt",
|
| 32 |
+
"chat_template.jinja", "special_tokens_map.json", "added_tokens.json", "generation_config.json",
|
| 33 |
+
"tokenizer.model", "preprocessor_config.json", "processor_config.json",
|
| 34 |
+
"video_preprocessor_config.json")
|
| 35 |
+
|
| 36 |
+
|
| 37 |
+
def load_remapper(path):
|
| 38 |
+
spec = importlib.util.spec_from_file_location("native_export_weight_mapping", path)
|
| 39 |
+
module = importlib.util.module_from_spec(spec)
|
| 40 |
+
spec.loader.exec_module(module)
|
| 41 |
+
return module.remap_qwen36_state_dict
|
| 42 |
+
|
| 43 |
+
|
| 44 |
+
def weight_error(original, restored, importance, module, torch):
|
| 45 |
+
"""Bounded row chunks; diagonal input second moments, not output KL."""
|
| 46 |
+
moments, count = None, None
|
| 47 |
+
if importance is not None and module + ".sumsq" in importance.files:
|
| 48 |
+
import numpy as np
|
| 49 |
+
|
| 50 |
+
sumsq = importance[module + ".sumsq"]
|
| 51 |
+
count = int(importance[module + ".count"].item())
|
| 52 |
+
if count < 1 or sumsq.shape != (original.shape[1],) or not np.isfinite(sumsq).all() or (sumsq < 0).any():
|
| 53 |
+
raise ValueError(f"Invalid activation importance for {module}")
|
| 54 |
+
moments = torch.from_numpy(sumsq.copy()).to(torch.float64) / count
|
| 55 |
+
error_sum, original_sum, weighted_error, weighted_signal, max_error = 0.0, 0.0, 0.0, 0.0, 0.0
|
| 56 |
+
for start in range(0, original.shape[0], 128):
|
| 57 |
+
source = original[start:start + 128].to(torch.float32)
|
| 58 |
+
error = restored[start:start + 128].to(torch.float32) - source
|
| 59 |
+
squared = error.square()
|
| 60 |
+
error_sum += squared.sum(dtype=torch.float64).item()
|
| 61 |
+
original_sum += source.square().sum(dtype=torch.float64).item()
|
| 62 |
+
max_error = max(max_error, error.abs().max().item())
|
| 63 |
+
if moments is not None:
|
| 64 |
+
weighted_error += (squared.sum(dim=0, dtype=torch.float64) * moments).sum().item()
|
| 65 |
+
weighted_signal += (source.square().sum(dim=0, dtype=torch.float64) * moments).sum().item()
|
| 66 |
+
result = {"sum_squared_error": error_sum, "mean_squared_error": error_sum / original.numel(),
|
| 67 |
+
"max_abs_error": max_error,
|
| 68 |
+
"relative_squared_error": error_sum / original_sum if original_sum else None,
|
| 69 |
+
"importance_available": moments is not None}
|
| 70 |
+
if moments is not None:
|
| 71 |
+
result.update({"activation_rows": count, "importance_weighted_squared_error": weighted_error,
|
| 72 |
+
"importance_weighted_mean_per_output": weighted_error / original.shape[0],
|
| 73 |
+
"importance_weighted_relative_squared_error": weighted_error / weighted_signal if weighted_signal else None})
|
| 74 |
+
return result
|
| 75 |
+
|
| 76 |
+
|
| 77 |
+
def export(args):
|
| 78 |
+
import numpy as np
|
| 79 |
+
import torch
|
| 80 |
+
import ttnn
|
| 81 |
+
from safetensors import safe_open
|
| 82 |
+
from safetensors.torch import save_file
|
| 83 |
+
from tt_eval import precision_map
|
| 84 |
+
|
| 85 |
+
torch.set_num_threads(args.cpu_threads)
|
| 86 |
+
weights, output = args.weights.resolve(), args.output.resolve()
|
| 87 |
+
if not output.is_relative_to(ROOT) or output == ROOT or output.is_relative_to(weights) or weights.is_relative_to(output):
|
| 88 |
+
raise ValueError(f"Output must be isolated beneath {ROOT}, outside original weights")
|
| 89 |
+
if output.exists() and any(output.iterdir()):
|
| 90 |
+
raise ValueError("Refusing to overwrite a nonempty output directory")
|
| 91 |
+
config = json.loads((weights / "config.json").read_text())
|
| 92 |
+
text = config["text_config"]
|
| 93 |
+
if (text["num_hidden_layers"], text["hidden_size"], text["vocab_size"]) != (32, 4096, 248320):
|
| 94 |
+
raise ValueError("Only original Qwen3.5-9B is supported")
|
| 95 |
+
overrides = json.loads(args.overrides.read_text()) if args.overrides else None
|
| 96 |
+
precision = precision_map(args.profile, overrides, text["layer_types"])
|
| 97 |
+
remap_path = ROOT / "runtime/qwen36/tt/weight_mapping.py"
|
| 98 |
+
remap = load_remapper(remap_path)
|
| 99 |
+
index_path = weights / "model.safetensors.index.json"
|
| 100 |
+
index = json.loads(index_path.read_text())["weight_map"]
|
| 101 |
+
validate_mtp_keys(index)
|
| 102 |
+
shards = {}
|
| 103 |
+
for name, filename in index.items():
|
| 104 |
+
if name.startswith("mtp.") or name == "lm_head.weight" or (name.startswith("model.language_model.") and ".mtp." not in name and ".visual." not in name):
|
| 105 |
+
shards.setdefault(filename, []).append(name)
|
| 106 |
+
if not shards:
|
| 107 |
+
raise ValueError("Expected original model.language_model text checkpoint keys")
|
| 108 |
+
importance = np.load(args.importance, allow_pickle=False) if args.importance else None
|
| 109 |
+
output.mkdir(parents=True, exist_ok=True)
|
| 110 |
+
(output / "tensors").mkdir()
|
| 111 |
+
(output / "provenance").mkdir()
|
| 112 |
+
files, tensors, source_shards, errors = {}, {}, {}, {}
|
| 113 |
+
|
| 114 |
+
def register(path):
|
| 115 |
+
path.chmod(0o644)
|
| 116 |
+
name = str(path.relative_to(output))
|
| 117 |
+
files[name] = {"bytes": path.stat().st_size, "sha256": digest(path)}
|
| 118 |
+
return name
|
| 119 |
+
|
| 120 |
+
for name in ASSETS:
|
| 121 |
+
source = weights / name
|
| 122 |
+
if source.is_file():
|
| 123 |
+
shutil.copyfile(source, output / name)
|
| 124 |
+
register(output / name)
|
| 125 |
+
licenses = [p for p in weights.iterdir() if p.is_file() and p.name.upper().startswith(("LICENSE", "NOTICE", "COPYING"))]
|
| 126 |
+
if not licenses or not all((output / name).is_file() for name in ("config.json", "tokenizer.json", "tokenizer_config.json")):
|
| 127 |
+
raise ValueError("Original config/tokenizer/license assets are required for standalone export")
|
| 128 |
+
for source in licenses:
|
| 129 |
+
shutil.copyfile(source, output / source.name)
|
| 130 |
+
register(output / source.name)
|
| 131 |
+
for source, destination in ((Path(__file__), output / "provenance/build_native_checkpoint.py"),
|
| 132 |
+
(ROOT / "native_checkpoint.py", output / "native_checkpoint.py"),
|
| 133 |
+
(remap_path, output / "provenance/weight_mapping.py"),
|
| 134 |
+
(ROOT / "tt_eval.py", output / "provenance/tt_eval.py"),
|
| 135 |
+
(ROOT / "precision-plan.json", output / "provenance/precision-plan.json"),
|
| 136 |
+
(index_path, output / "provenance/original.safetensors.index.json")):
|
| 137 |
+
shutil.copyfile(source, destination)
|
| 138 |
+
register(destination)
|
| 139 |
+
if args.overrides:
|
| 140 |
+
shutil.copyfile(args.overrides, output / "provenance/overrides.json")
|
| 141 |
+
register(output / "provenance/overrides.json")
|
| 142 |
+
if args.importance:
|
| 143 |
+
shutil.copyfile(args.importance, output / "provenance/importance.npz")
|
| 144 |
+
register(output / "provenance/importance.npz")
|
| 145 |
+
if (args.importance.parent / "metadata.json").is_file():
|
| 146 |
+
shutil.copyfile(args.importance.parent / "metadata.json", output / "provenance/importance-metadata.json")
|
| 147 |
+
register(output / "provenance/importance-metadata.json")
|
| 148 |
+
counter, quantized, fp32, mtp_fp32 = 0, 0, 0, 0
|
| 149 |
+
for filename, names in sorted(shards.items()):
|
| 150 |
+
shard = local_file(weights, filename)
|
| 151 |
+
source_shards[filename] = {"bytes": shard.stat().st_size, "sha256": digest(shard)}
|
| 152 |
+
with safe_open(str(shard), framework="pt", device="cpu") as source:
|
| 153 |
+
for original_name in sorted(names):
|
| 154 |
+
original = source.get_tensor(original_name)
|
| 155 |
+
original_hash = tensor_hash(original)
|
| 156 |
+
is_mtp = original_name.startswith("mtp.")
|
| 157 |
+
remapped = {original_name: original} if is_mtp else remap({original_name: original})
|
| 158 |
+
for name, value in remapped.items():
|
| 159 |
+
if name in tensors or "visual" in name or ("mtp" in name and not is_mtp):
|
| 160 |
+
raise ValueError(f"Unexpected duplicate/non-text remapped tensor: {name}")
|
| 161 |
+
dtype_name = None if is_mtp else tensor_precision(name, precision)
|
| 162 |
+
entry = {"source_name": original_name, "source_shard": filename,
|
| 163 |
+
"source_shape": list(original.shape), "source_torch_dtype": str(original.dtype),
|
| 164 |
+
"source_tensor_sha256": original_hash, "shape": list(value.shape),
|
| 165 |
+
"family": matrix_family(name), "restored_torch_dtype": str(value.dtype)}
|
| 166 |
+
if dtype_name is None:
|
| 167 |
+
value_hash = original_hash if is_mtp else tensor_hash(value)
|
| 168 |
+
path = output / "tensors" / f"{counter:05d}.safetensors"
|
| 169 |
+
save_file({name: value.contiguous()}, str(path))
|
| 170 |
+
with safe_open(str(path), framework="pt", device="cpu") as saved:
|
| 171 |
+
restored = saved.get_tensor(name)
|
| 172 |
+
if (restored.dtype != value.dtype or not torch.equal(restored, value)
|
| 173 |
+
or (is_mtp and tensor_hash(restored) != original_hash)):
|
| 174 |
+
raise ValueError(f"Lossless roundtrip failed: {name}")
|
| 175 |
+
del restored
|
| 176 |
+
if is_mtp:
|
| 177 |
+
mtp_fp32 += int(value.dtype == torch.float32)
|
| 178 |
+
else:
|
| 179 |
+
fp32 += int(value.dtype == torch.float32)
|
| 180 |
+
entry.update({"storage": "safetensors-lossless", "tensor_sha256": value_hash,
|
| 181 |
+
"roundtrip": {"exact_values": True, "exact_dtype": True}})
|
| 182 |
+
else:
|
| 183 |
+
if value.ndim != 2 or value.dtype != torch.bfloat16:
|
| 184 |
+
raise ValueError(f"Expected original BF16 linear matrix: {name} {value.shape} {value.dtype}")
|
| 185 |
+
path = output / "tensors" / f"{counter:05d}.tensorbin"
|
| 186 |
+
oriented = value.T.contiguous()
|
| 187 |
+
native = ttnn.from_torch(oriented, dtype=getattr(ttnn, DTYPES[dtype_name]), layout=ttnn.TILE_LAYOUT)
|
| 188 |
+
del oriented
|
| 189 |
+
ttnn.dump_tensor(str(path), native)
|
| 190 |
+
reloaded = ttnn.load_tensor(str(path))
|
| 191 |
+
rounded = ttnn.to_torch(reloaded).to(torch.bfloat16)
|
| 192 |
+
if not torch.equal(ttnn.to_torch(native), rounded):
|
| 193 |
+
raise ValueError(f"Native serialization or BF16 host restoration changes values: {name}")
|
| 194 |
+
del native, reloaded
|
| 195 |
+
restored = rounded.T.contiguous()
|
| 196 |
+
# Deliberately exercise the same transpose/copy that runtime converters do.
|
| 197 |
+
second = ttnn.from_torch(restored.T.contiguous(), dtype=getattr(ttnn, DTYPES[dtype_name]), layout=ttnn.TILE_LAYOUT)
|
| 198 |
+
second_path = output / "tensors" / f"{counter:05d}.roundtrip.tensorbin"
|
| 199 |
+
ttnn.dump_tensor(str(second_path), second)
|
| 200 |
+
exact_values = torch.equal(rounded, ttnn.to_torch(second))
|
| 201 |
+
exact_bytes = digest(path) == digest(second_path)
|
| 202 |
+
if not exact_values or not exact_bytes:
|
| 203 |
+
raise ValueError(f"Native requantization is not idempotent: {name}; values={exact_values}, bytes={exact_bytes}")
|
| 204 |
+
second_path.unlink()
|
| 205 |
+
del second, rounded
|
| 206 |
+
errors[name] = {"source_module": original_name.removesuffix(".weight"), "precision": dtype_name,
|
| 207 |
+
**weight_error(value, restored, importance, original_name.removesuffix(".weight"), torch)}
|
| 208 |
+
del restored
|
| 209 |
+
entry.update({"storage": "ttnn-tile", "precision": dtype_name,
|
| 210 |
+
"native_dtype": DTYPES[dtype_name], "native_shape": [value.shape[1], value.shape[0]],
|
| 211 |
+
"orientation": "input,output", "layout": "TILE_LAYOUT",
|
| 212 |
+
"roundtrip": {"exact_values": exact_values, "exact_serialized_bytes": exact_bytes}})
|
| 213 |
+
quantized += 1
|
| 214 |
+
entry["file"] = register(path)
|
| 215 |
+
entry["sha256"] = files[entry["file"]]["sha256"]
|
| 216 |
+
tensors[name] = entry
|
| 217 |
+
counter += 1
|
| 218 |
+
print(json.dumps({"tensor": name, "precision": dtype_name or "lossless", "bytes": files[entry["file"]]["bytes"]}), flush=True)
|
| 219 |
+
del original, remapped, value
|
| 220 |
+
if importance is not None:
|
| 221 |
+
importance.close()
|
| 222 |
+
if not {"tok_embeddings.weight", "output.weight", "norm.weight"} <= tensors.keys() or fp32 != 48:
|
| 223 |
+
raise ValueError(f"Missing top-level tensors or original FP32 nonlinear tensors: fp32={fp32}, expected 48")
|
| 224 |
+
validate_mtp_keys(tensors)
|
| 225 |
+
save_json(output / "quantization-error.json", {"method": "Unmodified TTNN rounding; importance-weighted precision ranking only, not optimized values or output KL",
|
| 226 |
+
"formula": "sum_out,in ((W-Wq)^2 * input_sumsq[in]/input_count)",
|
| 227 |
+
"tensors": errors})
|
| 228 |
+
register(output / "quantization-error.json")
|
| 229 |
+
manifest = {"format": FORMAT, "schema_version": 1, "scope": "text-only-no-vision-with-mtp",
|
| 230 |
+
"mtp": {"enabled": True, "num_speculative_tokens": 1,
|
| 231 |
+
"source_tensor_count": len(MTP_KEYS), "storage": "lossless"},
|
| 232 |
+
"status": "tensor_roundtrip_verified", "precision": precision,
|
| 233 |
+
"end_to_end_verification": "Requires separate manifest-bound equivalence.json (MTP=0) or equivalence-mtp.json (MTP=1); serialization alone is not validation",
|
| 234 |
+
"profile": args.profile, "files": files, "tensors": tensors,
|
| 235 |
+
"roundtrip": {"all_passed": True, "native_matrices": quantized, "lossless_tensors": counter - quantized,
|
| 236 |
+
"preserved_original_fp32_tensors": fp32,
|
| 237 |
+
"preserved_original_mtp_fp32_tensors": mtp_fp32},
|
| 238 |
+
"source": {"original_index_sha256": digest(index_path), "shards": source_shards,
|
| 239 |
+
"declared_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a"},
|
| 240 |
+
"toolchain": {"python": platform.python_version(), "torch": torch.__version__,
|
| 241 |
+
"ttnn": getattr(ttnn, "__version__", None), "container_image": args.container_image,
|
| 242 |
+
"tt_metal_commit": "de59f8a658b1ceafd230c8266026b1a72bb198d7"},
|
| 243 |
+
"load_contract": {"weights": "native dump -> CPU TT -> BF16 torch [out,in] -> runtime transpose -> requantize",
|
| 244 |
+
"nonlinear": "Lossless original tensors, preserving FP32 until normal runtime conversion",
|
| 245 |
+
"mtp": "Lossless original mtp.* tensors; existing Qwen36MTP uses target layer-0 precision policy with one speculative token",
|
| 246 |
+
"gdn_derived": "Runtime rebuilds AB and QKVABZ from quant-rounded components, then requantizes; no derived tensor stored",
|
| 247 |
+
"supported_runtime": "single-device Qwen36 text with optional MTP-1 at manifest precision, generic or explicitly dtype-gated packed families, only under verified runtime sources/environment",
|
| 248 |
+
"risks": ["Block exponent grouping is orientation/layout dependent", "Component idempotence does not prove GDN derived or full-model parity", "Every generic/packed configuration requires its own equivalence evidence; TP is excluded"],
|
| 249 |
+
"required_evidence": "native_checkpoint.py CHECKPOINT --baseline ORIGINAL_AT_SAME_PRECISION --restored NATIVE_RELOAD_RUN writes equivalence.json for recorded MTP=0 or equivalence-mtp.json for recorded MTP=1 only after exact full-logit comparison; actual speculation requires separate live-cycle evidence"},
|
| 250 |
+
"created_at": datetime.datetime.now(datetime.timezone.utc).isoformat()}
|
| 251 |
+
save_json(output / MANIFEST, manifest)
|
| 252 |
+
print(json.dumps({"manifest": str(output / MANIFEST), "tensors": counter, "native_matrices": quantized,
|
| 253 |
+
"artifact_bytes": sum(info["bytes"] for info in files.values()), "end_to_end_verified": False}), flush=True)
|
| 254 |
+
|
| 255 |
+
|
| 256 |
+
def main():
|
| 257 |
+
from tt_eval import PROFILES
|
| 258 |
+
|
| 259 |
+
parser = argparse.ArgumentParser(description=__doc__)
|
| 260 |
+
parser.add_argument("--weights", type=Path, required=True)
|
| 261 |
+
parser.add_argument("--output", type=Path, required=True)
|
| 262 |
+
parser.add_argument("--profile", choices=tuple(PROFILES), default="current-bfp4")
|
| 263 |
+
parser.add_argument("--overrides", type=Path)
|
| 264 |
+
parser.add_argument("--importance", type=Path)
|
| 265 |
+
parser.add_argument("--cpu-threads", type=int, default=8)
|
| 266 |
+
parser.add_argument("--container-image", required=True, help="Exact image identity reported by docker image inspect")
|
| 267 |
+
parser.add_argument("--device-ownership-confirmed", action="store_true")
|
| 268 |
+
args = parser.parse_args()
|
| 269 |
+
if not args.device_ownership_confirmed:
|
| 270 |
+
parser.error("TTNN host conversion requires exclusive device ownership; stop other P150 workloads first")
|
| 271 |
+
if args.cpu_threads < 1:
|
| 272 |
+
parser.error("--cpu-threads must be positive")
|
| 273 |
+
export(args)
|
| 274 |
+
|
| 275 |
+
|
| 276 |
+
if __name__ == "__main__":
|
| 277 |
+
main()
|
checkpoint/provenance/importance-metadata.json
ADDED
|
@@ -0,0 +1,1337 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"arguments": {
|
| 3 |
+
"accumulation_chunk_rows": 128,
|
| 4 |
+
"checkpoint_every": 32,
|
| 5 |
+
"input": "/work/instruction-calibration.jsonl",
|
| 6 |
+
"interop_threads": 1,
|
| 7 |
+
"max_tokens": 2048,
|
| 8 |
+
"model": "/model",
|
| 9 |
+
"output": "/work/importance",
|
| 10 |
+
"revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 11 |
+
"sample_rows": 8,
|
| 12 |
+
"threads": 16
|
| 13 |
+
},
|
| 14 |
+
"completed_input_tokens": 300941,
|
| 15 |
+
"completed_records": 614,
|
| 16 |
+
"config_sha256": "d0883072e01861ed0b2d47be3c16c36a8e81c224c7ffaa310c6558fb3f932b05",
|
| 17 |
+
"elapsed_seconds": 2298.5781133160344,
|
| 18 |
+
"exclusions": "Only torch.nn.Linear input channels. Embedding lookup, GDN convolution, norms and A_log are not diagonal-linear importance entries; original reference weights remain unchanged.",
|
| 19 |
+
"importance_sha256": "ca51deb3c0f6b096bea1ff2312b1bab4a4a8fa083ba0e03b282ec4f8ef303779",
|
| 20 |
+
"index_sha256": "26d3539b516be613f39563617cb9d33b3f83d401298125be392c80cefb8f7fe5",
|
| 21 |
+
"input_sha256": "c65c34c37633381a7322ff433a292f9e615d414bb834d1951b243ee16ecb310e",
|
| 22 |
+
"kind": "reference_cpu_linear_input_diagonal_second_moments",
|
| 23 |
+
"last_record_id": "instruct-multilingual-train-145250",
|
| 24 |
+
"loading_info": {
|
| 25 |
+
"error_msgs": [],
|
| 26 |
+
"mismatched_keys": [],
|
| 27 |
+
"missing_keys": [],
|
| 28 |
+
"unexpected_keys": []
|
| 29 |
+
},
|
| 30 |
+
"memory_policy": "One batch-size-1 complete record <= max-tokens; CPU BF16 reference with original FP32 GDN weights; use_cache=False; no vocabulary projection/logits; reduction temp <= accumulation-chunk-rows * largest_linear_input_features FP32; persistent float64 channel sums and optional sample-rows FP32 activation rows per module.",
|
| 31 |
+
"model_id": "Qwen/Qwen3.5-9B",
|
| 32 |
+
"model_path": "/model",
|
| 33 |
+
"model_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 34 |
+
"module_key_format": "Original checkpoint module name (strip .weight), keys <module>.sumsq float64[input_features], <module>.count int64 scalar; optional <module>.samples float32[rows,input_features] and <module>.sample_token_indices int64[rows]",
|
| 35 |
+
"modules": {
|
| 36 |
+
"lm_head": {
|
| 37 |
+
"count": 300941,
|
| 38 |
+
"input_features": 4096,
|
| 39 |
+
"sample_rows": 8
|
| 40 |
+
},
|
| 41 |
+
"model.language_model.layers.0.linear_attn.in_proj_a": {
|
| 42 |
+
"count": 300941,
|
| 43 |
+
"input_features": 4096,
|
| 44 |
+
"sample_rows": 8
|
| 45 |
+
},
|
| 46 |
+
"model.language_model.layers.0.linear_attn.in_proj_b": {
|
| 47 |
+
"count": 300941,
|
| 48 |
+
"input_features": 4096,
|
| 49 |
+
"sample_rows": 8
|
| 50 |
+
},
|
| 51 |
+
"model.language_model.layers.0.linear_attn.in_proj_qkv": {
|
| 52 |
+
"count": 300941,
|
| 53 |
+
"input_features": 4096,
|
| 54 |
+
"sample_rows": 8
|
| 55 |
+
},
|
| 56 |
+
"model.language_model.layers.0.linear_attn.in_proj_z": {
|
| 57 |
+
"count": 300941,
|
| 58 |
+
"input_features": 4096,
|
| 59 |
+
"sample_rows": 8
|
| 60 |
+
},
|
| 61 |
+
"model.language_model.layers.0.linear_attn.out_proj": {
|
| 62 |
+
"count": 300941,
|
| 63 |
+
"input_features": 4096,
|
| 64 |
+
"sample_rows": 8
|
| 65 |
+
},
|
| 66 |
+
"model.language_model.layers.0.mlp.down_proj": {
|
| 67 |
+
"count": 300941,
|
| 68 |
+
"input_features": 12288,
|
| 69 |
+
"sample_rows": 8
|
| 70 |
+
},
|
| 71 |
+
"model.language_model.layers.0.mlp.gate_proj": {
|
| 72 |
+
"count": 300941,
|
| 73 |
+
"input_features": 4096,
|
| 74 |
+
"sample_rows": 8
|
| 75 |
+
},
|
| 76 |
+
"model.language_model.layers.0.mlp.up_proj": {
|
| 77 |
+
"count": 300941,
|
| 78 |
+
"input_features": 4096,
|
| 79 |
+
"sample_rows": 8
|
| 80 |
+
},
|
| 81 |
+
"model.language_model.layers.1.linear_attn.in_proj_a": {
|
| 82 |
+
"count": 300941,
|
| 83 |
+
"input_features": 4096,
|
| 84 |
+
"sample_rows": 8
|
| 85 |
+
},
|
| 86 |
+
"model.language_model.layers.1.linear_attn.in_proj_b": {
|
| 87 |
+
"count": 300941,
|
| 88 |
+
"input_features": 4096,
|
| 89 |
+
"sample_rows": 8
|
| 90 |
+
},
|
| 91 |
+
"model.language_model.layers.1.linear_attn.in_proj_qkv": {
|
| 92 |
+
"count": 300941,
|
| 93 |
+
"input_features": 4096,
|
| 94 |
+
"sample_rows": 8
|
| 95 |
+
},
|
| 96 |
+
"model.language_model.layers.1.linear_attn.in_proj_z": {
|
| 97 |
+
"count": 300941,
|
| 98 |
+
"input_features": 4096,
|
| 99 |
+
"sample_rows": 8
|
| 100 |
+
},
|
| 101 |
+
"model.language_model.layers.1.linear_attn.out_proj": {
|
| 102 |
+
"count": 300941,
|
| 103 |
+
"input_features": 4096,
|
| 104 |
+
"sample_rows": 8
|
| 105 |
+
},
|
| 106 |
+
"model.language_model.layers.1.mlp.down_proj": {
|
| 107 |
+
"count": 300941,
|
| 108 |
+
"input_features": 12288,
|
| 109 |
+
"sample_rows": 8
|
| 110 |
+
},
|
| 111 |
+
"model.language_model.layers.1.mlp.gate_proj": {
|
| 112 |
+
"count": 300941,
|
| 113 |
+
"input_features": 4096,
|
| 114 |
+
"sample_rows": 8
|
| 115 |
+
},
|
| 116 |
+
"model.language_model.layers.1.mlp.up_proj": {
|
| 117 |
+
"count": 300941,
|
| 118 |
+
"input_features": 4096,
|
| 119 |
+
"sample_rows": 8
|
| 120 |
+
},
|
| 121 |
+
"model.language_model.layers.10.linear_attn.in_proj_a": {
|
| 122 |
+
"count": 300941,
|
| 123 |
+
"input_features": 4096,
|
| 124 |
+
"sample_rows": 8
|
| 125 |
+
},
|
| 126 |
+
"model.language_model.layers.10.linear_attn.in_proj_b": {
|
| 127 |
+
"count": 300941,
|
| 128 |
+
"input_features": 4096,
|
| 129 |
+
"sample_rows": 8
|
| 130 |
+
},
|
| 131 |
+
"model.language_model.layers.10.linear_attn.in_proj_qkv": {
|
| 132 |
+
"count": 300941,
|
| 133 |
+
"input_features": 4096,
|
| 134 |
+
"sample_rows": 8
|
| 135 |
+
},
|
| 136 |
+
"model.language_model.layers.10.linear_attn.in_proj_z": {
|
| 137 |
+
"count": 300941,
|
| 138 |
+
"input_features": 4096,
|
| 139 |
+
"sample_rows": 8
|
| 140 |
+
},
|
| 141 |
+
"model.language_model.layers.10.linear_attn.out_proj": {
|
| 142 |
+
"count": 300941,
|
| 143 |
+
"input_features": 4096,
|
| 144 |
+
"sample_rows": 8
|
| 145 |
+
},
|
| 146 |
+
"model.language_model.layers.10.mlp.down_proj": {
|
| 147 |
+
"count": 300941,
|
| 148 |
+
"input_features": 12288,
|
| 149 |
+
"sample_rows": 8
|
| 150 |
+
},
|
| 151 |
+
"model.language_model.layers.10.mlp.gate_proj": {
|
| 152 |
+
"count": 300941,
|
| 153 |
+
"input_features": 4096,
|
| 154 |
+
"sample_rows": 8
|
| 155 |
+
},
|
| 156 |
+
"model.language_model.layers.10.mlp.up_proj": {
|
| 157 |
+
"count": 300941,
|
| 158 |
+
"input_features": 4096,
|
| 159 |
+
"sample_rows": 8
|
| 160 |
+
},
|
| 161 |
+
"model.language_model.layers.11.mlp.down_proj": {
|
| 162 |
+
"count": 300941,
|
| 163 |
+
"input_features": 12288,
|
| 164 |
+
"sample_rows": 8
|
| 165 |
+
},
|
| 166 |
+
"model.language_model.layers.11.mlp.gate_proj": {
|
| 167 |
+
"count": 300941,
|
| 168 |
+
"input_features": 4096,
|
| 169 |
+
"sample_rows": 8
|
| 170 |
+
},
|
| 171 |
+
"model.language_model.layers.11.mlp.up_proj": {
|
| 172 |
+
"count": 300941,
|
| 173 |
+
"input_features": 4096,
|
| 174 |
+
"sample_rows": 8
|
| 175 |
+
},
|
| 176 |
+
"model.language_model.layers.11.self_attn.k_proj": {
|
| 177 |
+
"count": 300941,
|
| 178 |
+
"input_features": 4096,
|
| 179 |
+
"sample_rows": 8
|
| 180 |
+
},
|
| 181 |
+
"model.language_model.layers.11.self_attn.o_proj": {
|
| 182 |
+
"count": 300941,
|
| 183 |
+
"input_features": 4096,
|
| 184 |
+
"sample_rows": 8
|
| 185 |
+
},
|
| 186 |
+
"model.language_model.layers.11.self_attn.q_proj": {
|
| 187 |
+
"count": 300941,
|
| 188 |
+
"input_features": 4096,
|
| 189 |
+
"sample_rows": 8
|
| 190 |
+
},
|
| 191 |
+
"model.language_model.layers.11.self_attn.v_proj": {
|
| 192 |
+
"count": 300941,
|
| 193 |
+
"input_features": 4096,
|
| 194 |
+
"sample_rows": 8
|
| 195 |
+
},
|
| 196 |
+
"model.language_model.layers.12.linear_attn.in_proj_a": {
|
| 197 |
+
"count": 300941,
|
| 198 |
+
"input_features": 4096,
|
| 199 |
+
"sample_rows": 8
|
| 200 |
+
},
|
| 201 |
+
"model.language_model.layers.12.linear_attn.in_proj_b": {
|
| 202 |
+
"count": 300941,
|
| 203 |
+
"input_features": 4096,
|
| 204 |
+
"sample_rows": 8
|
| 205 |
+
},
|
| 206 |
+
"model.language_model.layers.12.linear_attn.in_proj_qkv": {
|
| 207 |
+
"count": 300941,
|
| 208 |
+
"input_features": 4096,
|
| 209 |
+
"sample_rows": 8
|
| 210 |
+
},
|
| 211 |
+
"model.language_model.layers.12.linear_attn.in_proj_z": {
|
| 212 |
+
"count": 300941,
|
| 213 |
+
"input_features": 4096,
|
| 214 |
+
"sample_rows": 8
|
| 215 |
+
},
|
| 216 |
+
"model.language_model.layers.12.linear_attn.out_proj": {
|
| 217 |
+
"count": 300941,
|
| 218 |
+
"input_features": 4096,
|
| 219 |
+
"sample_rows": 8
|
| 220 |
+
},
|
| 221 |
+
"model.language_model.layers.12.mlp.down_proj": {
|
| 222 |
+
"count": 300941,
|
| 223 |
+
"input_features": 12288,
|
| 224 |
+
"sample_rows": 8
|
| 225 |
+
},
|
| 226 |
+
"model.language_model.layers.12.mlp.gate_proj": {
|
| 227 |
+
"count": 300941,
|
| 228 |
+
"input_features": 4096,
|
| 229 |
+
"sample_rows": 8
|
| 230 |
+
},
|
| 231 |
+
"model.language_model.layers.12.mlp.up_proj": {
|
| 232 |
+
"count": 300941,
|
| 233 |
+
"input_features": 4096,
|
| 234 |
+
"sample_rows": 8
|
| 235 |
+
},
|
| 236 |
+
"model.language_model.layers.13.linear_attn.in_proj_a": {
|
| 237 |
+
"count": 300941,
|
| 238 |
+
"input_features": 4096,
|
| 239 |
+
"sample_rows": 8
|
| 240 |
+
},
|
| 241 |
+
"model.language_model.layers.13.linear_attn.in_proj_b": {
|
| 242 |
+
"count": 300941,
|
| 243 |
+
"input_features": 4096,
|
| 244 |
+
"sample_rows": 8
|
| 245 |
+
},
|
| 246 |
+
"model.language_model.layers.13.linear_attn.in_proj_qkv": {
|
| 247 |
+
"count": 300941,
|
| 248 |
+
"input_features": 4096,
|
| 249 |
+
"sample_rows": 8
|
| 250 |
+
},
|
| 251 |
+
"model.language_model.layers.13.linear_attn.in_proj_z": {
|
| 252 |
+
"count": 300941,
|
| 253 |
+
"input_features": 4096,
|
| 254 |
+
"sample_rows": 8
|
| 255 |
+
},
|
| 256 |
+
"model.language_model.layers.13.linear_attn.out_proj": {
|
| 257 |
+
"count": 300941,
|
| 258 |
+
"input_features": 4096,
|
| 259 |
+
"sample_rows": 8
|
| 260 |
+
},
|
| 261 |
+
"model.language_model.layers.13.mlp.down_proj": {
|
| 262 |
+
"count": 300941,
|
| 263 |
+
"input_features": 12288,
|
| 264 |
+
"sample_rows": 8
|
| 265 |
+
},
|
| 266 |
+
"model.language_model.layers.13.mlp.gate_proj": {
|
| 267 |
+
"count": 300941,
|
| 268 |
+
"input_features": 4096,
|
| 269 |
+
"sample_rows": 8
|
| 270 |
+
},
|
| 271 |
+
"model.language_model.layers.13.mlp.up_proj": {
|
| 272 |
+
"count": 300941,
|
| 273 |
+
"input_features": 4096,
|
| 274 |
+
"sample_rows": 8
|
| 275 |
+
},
|
| 276 |
+
"model.language_model.layers.14.linear_attn.in_proj_a": {
|
| 277 |
+
"count": 300941,
|
| 278 |
+
"input_features": 4096,
|
| 279 |
+
"sample_rows": 8
|
| 280 |
+
},
|
| 281 |
+
"model.language_model.layers.14.linear_attn.in_proj_b": {
|
| 282 |
+
"count": 300941,
|
| 283 |
+
"input_features": 4096,
|
| 284 |
+
"sample_rows": 8
|
| 285 |
+
},
|
| 286 |
+
"model.language_model.layers.14.linear_attn.in_proj_qkv": {
|
| 287 |
+
"count": 300941,
|
| 288 |
+
"input_features": 4096,
|
| 289 |
+
"sample_rows": 8
|
| 290 |
+
},
|
| 291 |
+
"model.language_model.layers.14.linear_attn.in_proj_z": {
|
| 292 |
+
"count": 300941,
|
| 293 |
+
"input_features": 4096,
|
| 294 |
+
"sample_rows": 8
|
| 295 |
+
},
|
| 296 |
+
"model.language_model.layers.14.linear_attn.out_proj": {
|
| 297 |
+
"count": 300941,
|
| 298 |
+
"input_features": 4096,
|
| 299 |
+
"sample_rows": 8
|
| 300 |
+
},
|
| 301 |
+
"model.language_model.layers.14.mlp.down_proj": {
|
| 302 |
+
"count": 300941,
|
| 303 |
+
"input_features": 12288,
|
| 304 |
+
"sample_rows": 8
|
| 305 |
+
},
|
| 306 |
+
"model.language_model.layers.14.mlp.gate_proj": {
|
| 307 |
+
"count": 300941,
|
| 308 |
+
"input_features": 4096,
|
| 309 |
+
"sample_rows": 8
|
| 310 |
+
},
|
| 311 |
+
"model.language_model.layers.14.mlp.up_proj": {
|
| 312 |
+
"count": 300941,
|
| 313 |
+
"input_features": 4096,
|
| 314 |
+
"sample_rows": 8
|
| 315 |
+
},
|
| 316 |
+
"model.language_model.layers.15.mlp.down_proj": {
|
| 317 |
+
"count": 300941,
|
| 318 |
+
"input_features": 12288,
|
| 319 |
+
"sample_rows": 8
|
| 320 |
+
},
|
| 321 |
+
"model.language_model.layers.15.mlp.gate_proj": {
|
| 322 |
+
"count": 300941,
|
| 323 |
+
"input_features": 4096,
|
| 324 |
+
"sample_rows": 8
|
| 325 |
+
},
|
| 326 |
+
"model.language_model.layers.15.mlp.up_proj": {
|
| 327 |
+
"count": 300941,
|
| 328 |
+
"input_features": 4096,
|
| 329 |
+
"sample_rows": 8
|
| 330 |
+
},
|
| 331 |
+
"model.language_model.layers.15.self_attn.k_proj": {
|
| 332 |
+
"count": 300941,
|
| 333 |
+
"input_features": 4096,
|
| 334 |
+
"sample_rows": 8
|
| 335 |
+
},
|
| 336 |
+
"model.language_model.layers.15.self_attn.o_proj": {
|
| 337 |
+
"count": 300941,
|
| 338 |
+
"input_features": 4096,
|
| 339 |
+
"sample_rows": 8
|
| 340 |
+
},
|
| 341 |
+
"model.language_model.layers.15.self_attn.q_proj": {
|
| 342 |
+
"count": 300941,
|
| 343 |
+
"input_features": 4096,
|
| 344 |
+
"sample_rows": 8
|
| 345 |
+
},
|
| 346 |
+
"model.language_model.layers.15.self_attn.v_proj": {
|
| 347 |
+
"count": 300941,
|
| 348 |
+
"input_features": 4096,
|
| 349 |
+
"sample_rows": 8
|
| 350 |
+
},
|
| 351 |
+
"model.language_model.layers.16.linear_attn.in_proj_a": {
|
| 352 |
+
"count": 300941,
|
| 353 |
+
"input_features": 4096,
|
| 354 |
+
"sample_rows": 8
|
| 355 |
+
},
|
| 356 |
+
"model.language_model.layers.16.linear_attn.in_proj_b": {
|
| 357 |
+
"count": 300941,
|
| 358 |
+
"input_features": 4096,
|
| 359 |
+
"sample_rows": 8
|
| 360 |
+
},
|
| 361 |
+
"model.language_model.layers.16.linear_attn.in_proj_qkv": {
|
| 362 |
+
"count": 300941,
|
| 363 |
+
"input_features": 4096,
|
| 364 |
+
"sample_rows": 8
|
| 365 |
+
},
|
| 366 |
+
"model.language_model.layers.16.linear_attn.in_proj_z": {
|
| 367 |
+
"count": 300941,
|
| 368 |
+
"input_features": 4096,
|
| 369 |
+
"sample_rows": 8
|
| 370 |
+
},
|
| 371 |
+
"model.language_model.layers.16.linear_attn.out_proj": {
|
| 372 |
+
"count": 300941,
|
| 373 |
+
"input_features": 4096,
|
| 374 |
+
"sample_rows": 8
|
| 375 |
+
},
|
| 376 |
+
"model.language_model.layers.16.mlp.down_proj": {
|
| 377 |
+
"count": 300941,
|
| 378 |
+
"input_features": 12288,
|
| 379 |
+
"sample_rows": 8
|
| 380 |
+
},
|
| 381 |
+
"model.language_model.layers.16.mlp.gate_proj": {
|
| 382 |
+
"count": 300941,
|
| 383 |
+
"input_features": 4096,
|
| 384 |
+
"sample_rows": 8
|
| 385 |
+
},
|
| 386 |
+
"model.language_model.layers.16.mlp.up_proj": {
|
| 387 |
+
"count": 300941,
|
| 388 |
+
"input_features": 4096,
|
| 389 |
+
"sample_rows": 8
|
| 390 |
+
},
|
| 391 |
+
"model.language_model.layers.17.linear_attn.in_proj_a": {
|
| 392 |
+
"count": 300941,
|
| 393 |
+
"input_features": 4096,
|
| 394 |
+
"sample_rows": 8
|
| 395 |
+
},
|
| 396 |
+
"model.language_model.layers.17.linear_attn.in_proj_b": {
|
| 397 |
+
"count": 300941,
|
| 398 |
+
"input_features": 4096,
|
| 399 |
+
"sample_rows": 8
|
| 400 |
+
},
|
| 401 |
+
"model.language_model.layers.17.linear_attn.in_proj_qkv": {
|
| 402 |
+
"count": 300941,
|
| 403 |
+
"input_features": 4096,
|
| 404 |
+
"sample_rows": 8
|
| 405 |
+
},
|
| 406 |
+
"model.language_model.layers.17.linear_attn.in_proj_z": {
|
| 407 |
+
"count": 300941,
|
| 408 |
+
"input_features": 4096,
|
| 409 |
+
"sample_rows": 8
|
| 410 |
+
},
|
| 411 |
+
"model.language_model.layers.17.linear_attn.out_proj": {
|
| 412 |
+
"count": 300941,
|
| 413 |
+
"input_features": 4096,
|
| 414 |
+
"sample_rows": 8
|
| 415 |
+
},
|
| 416 |
+
"model.language_model.layers.17.mlp.down_proj": {
|
| 417 |
+
"count": 300941,
|
| 418 |
+
"input_features": 12288,
|
| 419 |
+
"sample_rows": 8
|
| 420 |
+
},
|
| 421 |
+
"model.language_model.layers.17.mlp.gate_proj": {
|
| 422 |
+
"count": 300941,
|
| 423 |
+
"input_features": 4096,
|
| 424 |
+
"sample_rows": 8
|
| 425 |
+
},
|
| 426 |
+
"model.language_model.layers.17.mlp.up_proj": {
|
| 427 |
+
"count": 300941,
|
| 428 |
+
"input_features": 4096,
|
| 429 |
+
"sample_rows": 8
|
| 430 |
+
},
|
| 431 |
+
"model.language_model.layers.18.linear_attn.in_proj_a": {
|
| 432 |
+
"count": 300941,
|
| 433 |
+
"input_features": 4096,
|
| 434 |
+
"sample_rows": 8
|
| 435 |
+
},
|
| 436 |
+
"model.language_model.layers.18.linear_attn.in_proj_b": {
|
| 437 |
+
"count": 300941,
|
| 438 |
+
"input_features": 4096,
|
| 439 |
+
"sample_rows": 8
|
| 440 |
+
},
|
| 441 |
+
"model.language_model.layers.18.linear_attn.in_proj_qkv": {
|
| 442 |
+
"count": 300941,
|
| 443 |
+
"input_features": 4096,
|
| 444 |
+
"sample_rows": 8
|
| 445 |
+
},
|
| 446 |
+
"model.language_model.layers.18.linear_attn.in_proj_z": {
|
| 447 |
+
"count": 300941,
|
| 448 |
+
"input_features": 4096,
|
| 449 |
+
"sample_rows": 8
|
| 450 |
+
},
|
| 451 |
+
"model.language_model.layers.18.linear_attn.out_proj": {
|
| 452 |
+
"count": 300941,
|
| 453 |
+
"input_features": 4096,
|
| 454 |
+
"sample_rows": 8
|
| 455 |
+
},
|
| 456 |
+
"model.language_model.layers.18.mlp.down_proj": {
|
| 457 |
+
"count": 300941,
|
| 458 |
+
"input_features": 12288,
|
| 459 |
+
"sample_rows": 8
|
| 460 |
+
},
|
| 461 |
+
"model.language_model.layers.18.mlp.gate_proj": {
|
| 462 |
+
"count": 300941,
|
| 463 |
+
"input_features": 4096,
|
| 464 |
+
"sample_rows": 8
|
| 465 |
+
},
|
| 466 |
+
"model.language_model.layers.18.mlp.up_proj": {
|
| 467 |
+
"count": 300941,
|
| 468 |
+
"input_features": 4096,
|
| 469 |
+
"sample_rows": 8
|
| 470 |
+
},
|
| 471 |
+
"model.language_model.layers.19.mlp.down_proj": {
|
| 472 |
+
"count": 300941,
|
| 473 |
+
"input_features": 12288,
|
| 474 |
+
"sample_rows": 8
|
| 475 |
+
},
|
| 476 |
+
"model.language_model.layers.19.mlp.gate_proj": {
|
| 477 |
+
"count": 300941,
|
| 478 |
+
"input_features": 4096,
|
| 479 |
+
"sample_rows": 8
|
| 480 |
+
},
|
| 481 |
+
"model.language_model.layers.19.mlp.up_proj": {
|
| 482 |
+
"count": 300941,
|
| 483 |
+
"input_features": 4096,
|
| 484 |
+
"sample_rows": 8
|
| 485 |
+
},
|
| 486 |
+
"model.language_model.layers.19.self_attn.k_proj": {
|
| 487 |
+
"count": 300941,
|
| 488 |
+
"input_features": 4096,
|
| 489 |
+
"sample_rows": 8
|
| 490 |
+
},
|
| 491 |
+
"model.language_model.layers.19.self_attn.o_proj": {
|
| 492 |
+
"count": 300941,
|
| 493 |
+
"input_features": 4096,
|
| 494 |
+
"sample_rows": 8
|
| 495 |
+
},
|
| 496 |
+
"model.language_model.layers.19.self_attn.q_proj": {
|
| 497 |
+
"count": 300941,
|
| 498 |
+
"input_features": 4096,
|
| 499 |
+
"sample_rows": 8
|
| 500 |
+
},
|
| 501 |
+
"model.language_model.layers.19.self_attn.v_proj": {
|
| 502 |
+
"count": 300941,
|
| 503 |
+
"input_features": 4096,
|
| 504 |
+
"sample_rows": 8
|
| 505 |
+
},
|
| 506 |
+
"model.language_model.layers.2.linear_attn.in_proj_a": {
|
| 507 |
+
"count": 300941,
|
| 508 |
+
"input_features": 4096,
|
| 509 |
+
"sample_rows": 8
|
| 510 |
+
},
|
| 511 |
+
"model.language_model.layers.2.linear_attn.in_proj_b": {
|
| 512 |
+
"count": 300941,
|
| 513 |
+
"input_features": 4096,
|
| 514 |
+
"sample_rows": 8
|
| 515 |
+
},
|
| 516 |
+
"model.language_model.layers.2.linear_attn.in_proj_qkv": {
|
| 517 |
+
"count": 300941,
|
| 518 |
+
"input_features": 4096,
|
| 519 |
+
"sample_rows": 8
|
| 520 |
+
},
|
| 521 |
+
"model.language_model.layers.2.linear_attn.in_proj_z": {
|
| 522 |
+
"count": 300941,
|
| 523 |
+
"input_features": 4096,
|
| 524 |
+
"sample_rows": 8
|
| 525 |
+
},
|
| 526 |
+
"model.language_model.layers.2.linear_attn.out_proj": {
|
| 527 |
+
"count": 300941,
|
| 528 |
+
"input_features": 4096,
|
| 529 |
+
"sample_rows": 8
|
| 530 |
+
},
|
| 531 |
+
"model.language_model.layers.2.mlp.down_proj": {
|
| 532 |
+
"count": 300941,
|
| 533 |
+
"input_features": 12288,
|
| 534 |
+
"sample_rows": 8
|
| 535 |
+
},
|
| 536 |
+
"model.language_model.layers.2.mlp.gate_proj": {
|
| 537 |
+
"count": 300941,
|
| 538 |
+
"input_features": 4096,
|
| 539 |
+
"sample_rows": 8
|
| 540 |
+
},
|
| 541 |
+
"model.language_model.layers.2.mlp.up_proj": {
|
| 542 |
+
"count": 300941,
|
| 543 |
+
"input_features": 4096,
|
| 544 |
+
"sample_rows": 8
|
| 545 |
+
},
|
| 546 |
+
"model.language_model.layers.20.linear_attn.in_proj_a": {
|
| 547 |
+
"count": 300941,
|
| 548 |
+
"input_features": 4096,
|
| 549 |
+
"sample_rows": 8
|
| 550 |
+
},
|
| 551 |
+
"model.language_model.layers.20.linear_attn.in_proj_b": {
|
| 552 |
+
"count": 300941,
|
| 553 |
+
"input_features": 4096,
|
| 554 |
+
"sample_rows": 8
|
| 555 |
+
},
|
| 556 |
+
"model.language_model.layers.20.linear_attn.in_proj_qkv": {
|
| 557 |
+
"count": 300941,
|
| 558 |
+
"input_features": 4096,
|
| 559 |
+
"sample_rows": 8
|
| 560 |
+
},
|
| 561 |
+
"model.language_model.layers.20.linear_attn.in_proj_z": {
|
| 562 |
+
"count": 300941,
|
| 563 |
+
"input_features": 4096,
|
| 564 |
+
"sample_rows": 8
|
| 565 |
+
},
|
| 566 |
+
"model.language_model.layers.20.linear_attn.out_proj": {
|
| 567 |
+
"count": 300941,
|
| 568 |
+
"input_features": 4096,
|
| 569 |
+
"sample_rows": 8
|
| 570 |
+
},
|
| 571 |
+
"model.language_model.layers.20.mlp.down_proj": {
|
| 572 |
+
"count": 300941,
|
| 573 |
+
"input_features": 12288,
|
| 574 |
+
"sample_rows": 8
|
| 575 |
+
},
|
| 576 |
+
"model.language_model.layers.20.mlp.gate_proj": {
|
| 577 |
+
"count": 300941,
|
| 578 |
+
"input_features": 4096,
|
| 579 |
+
"sample_rows": 8
|
| 580 |
+
},
|
| 581 |
+
"model.language_model.layers.20.mlp.up_proj": {
|
| 582 |
+
"count": 300941,
|
| 583 |
+
"input_features": 4096,
|
| 584 |
+
"sample_rows": 8
|
| 585 |
+
},
|
| 586 |
+
"model.language_model.layers.21.linear_attn.in_proj_a": {
|
| 587 |
+
"count": 300941,
|
| 588 |
+
"input_features": 4096,
|
| 589 |
+
"sample_rows": 8
|
| 590 |
+
},
|
| 591 |
+
"model.language_model.layers.21.linear_attn.in_proj_b": {
|
| 592 |
+
"count": 300941,
|
| 593 |
+
"input_features": 4096,
|
| 594 |
+
"sample_rows": 8
|
| 595 |
+
},
|
| 596 |
+
"model.language_model.layers.21.linear_attn.in_proj_qkv": {
|
| 597 |
+
"count": 300941,
|
| 598 |
+
"input_features": 4096,
|
| 599 |
+
"sample_rows": 8
|
| 600 |
+
},
|
| 601 |
+
"model.language_model.layers.21.linear_attn.in_proj_z": {
|
| 602 |
+
"count": 300941,
|
| 603 |
+
"input_features": 4096,
|
| 604 |
+
"sample_rows": 8
|
| 605 |
+
},
|
| 606 |
+
"model.language_model.layers.21.linear_attn.out_proj": {
|
| 607 |
+
"count": 300941,
|
| 608 |
+
"input_features": 4096,
|
| 609 |
+
"sample_rows": 8
|
| 610 |
+
},
|
| 611 |
+
"model.language_model.layers.21.mlp.down_proj": {
|
| 612 |
+
"count": 300941,
|
| 613 |
+
"input_features": 12288,
|
| 614 |
+
"sample_rows": 8
|
| 615 |
+
},
|
| 616 |
+
"model.language_model.layers.21.mlp.gate_proj": {
|
| 617 |
+
"count": 300941,
|
| 618 |
+
"input_features": 4096,
|
| 619 |
+
"sample_rows": 8
|
| 620 |
+
},
|
| 621 |
+
"model.language_model.layers.21.mlp.up_proj": {
|
| 622 |
+
"count": 300941,
|
| 623 |
+
"input_features": 4096,
|
| 624 |
+
"sample_rows": 8
|
| 625 |
+
},
|
| 626 |
+
"model.language_model.layers.22.linear_attn.in_proj_a": {
|
| 627 |
+
"count": 300941,
|
| 628 |
+
"input_features": 4096,
|
| 629 |
+
"sample_rows": 8
|
| 630 |
+
},
|
| 631 |
+
"model.language_model.layers.22.linear_attn.in_proj_b": {
|
| 632 |
+
"count": 300941,
|
| 633 |
+
"input_features": 4096,
|
| 634 |
+
"sample_rows": 8
|
| 635 |
+
},
|
| 636 |
+
"model.language_model.layers.22.linear_attn.in_proj_qkv": {
|
| 637 |
+
"count": 300941,
|
| 638 |
+
"input_features": 4096,
|
| 639 |
+
"sample_rows": 8
|
| 640 |
+
},
|
| 641 |
+
"model.language_model.layers.22.linear_attn.in_proj_z": {
|
| 642 |
+
"count": 300941,
|
| 643 |
+
"input_features": 4096,
|
| 644 |
+
"sample_rows": 8
|
| 645 |
+
},
|
| 646 |
+
"model.language_model.layers.22.linear_attn.out_proj": {
|
| 647 |
+
"count": 300941,
|
| 648 |
+
"input_features": 4096,
|
| 649 |
+
"sample_rows": 8
|
| 650 |
+
},
|
| 651 |
+
"model.language_model.layers.22.mlp.down_proj": {
|
| 652 |
+
"count": 300941,
|
| 653 |
+
"input_features": 12288,
|
| 654 |
+
"sample_rows": 8
|
| 655 |
+
},
|
| 656 |
+
"model.language_model.layers.22.mlp.gate_proj": {
|
| 657 |
+
"count": 300941,
|
| 658 |
+
"input_features": 4096,
|
| 659 |
+
"sample_rows": 8
|
| 660 |
+
},
|
| 661 |
+
"model.language_model.layers.22.mlp.up_proj": {
|
| 662 |
+
"count": 300941,
|
| 663 |
+
"input_features": 4096,
|
| 664 |
+
"sample_rows": 8
|
| 665 |
+
},
|
| 666 |
+
"model.language_model.layers.23.mlp.down_proj": {
|
| 667 |
+
"count": 300941,
|
| 668 |
+
"input_features": 12288,
|
| 669 |
+
"sample_rows": 8
|
| 670 |
+
},
|
| 671 |
+
"model.language_model.layers.23.mlp.gate_proj": {
|
| 672 |
+
"count": 300941,
|
| 673 |
+
"input_features": 4096,
|
| 674 |
+
"sample_rows": 8
|
| 675 |
+
},
|
| 676 |
+
"model.language_model.layers.23.mlp.up_proj": {
|
| 677 |
+
"count": 300941,
|
| 678 |
+
"input_features": 4096,
|
| 679 |
+
"sample_rows": 8
|
| 680 |
+
},
|
| 681 |
+
"model.language_model.layers.23.self_attn.k_proj": {
|
| 682 |
+
"count": 300941,
|
| 683 |
+
"input_features": 4096,
|
| 684 |
+
"sample_rows": 8
|
| 685 |
+
},
|
| 686 |
+
"model.language_model.layers.23.self_attn.o_proj": {
|
| 687 |
+
"count": 300941,
|
| 688 |
+
"input_features": 4096,
|
| 689 |
+
"sample_rows": 8
|
| 690 |
+
},
|
| 691 |
+
"model.language_model.layers.23.self_attn.q_proj": {
|
| 692 |
+
"count": 300941,
|
| 693 |
+
"input_features": 4096,
|
| 694 |
+
"sample_rows": 8
|
| 695 |
+
},
|
| 696 |
+
"model.language_model.layers.23.self_attn.v_proj": {
|
| 697 |
+
"count": 300941,
|
| 698 |
+
"input_features": 4096,
|
| 699 |
+
"sample_rows": 8
|
| 700 |
+
},
|
| 701 |
+
"model.language_model.layers.24.linear_attn.in_proj_a": {
|
| 702 |
+
"count": 300941,
|
| 703 |
+
"input_features": 4096,
|
| 704 |
+
"sample_rows": 8
|
| 705 |
+
},
|
| 706 |
+
"model.language_model.layers.24.linear_attn.in_proj_b": {
|
| 707 |
+
"count": 300941,
|
| 708 |
+
"input_features": 4096,
|
| 709 |
+
"sample_rows": 8
|
| 710 |
+
},
|
| 711 |
+
"model.language_model.layers.24.linear_attn.in_proj_qkv": {
|
| 712 |
+
"count": 300941,
|
| 713 |
+
"input_features": 4096,
|
| 714 |
+
"sample_rows": 8
|
| 715 |
+
},
|
| 716 |
+
"model.language_model.layers.24.linear_attn.in_proj_z": {
|
| 717 |
+
"count": 300941,
|
| 718 |
+
"input_features": 4096,
|
| 719 |
+
"sample_rows": 8
|
| 720 |
+
},
|
| 721 |
+
"model.language_model.layers.24.linear_attn.out_proj": {
|
| 722 |
+
"count": 300941,
|
| 723 |
+
"input_features": 4096,
|
| 724 |
+
"sample_rows": 8
|
| 725 |
+
},
|
| 726 |
+
"model.language_model.layers.24.mlp.down_proj": {
|
| 727 |
+
"count": 300941,
|
| 728 |
+
"input_features": 12288,
|
| 729 |
+
"sample_rows": 8
|
| 730 |
+
},
|
| 731 |
+
"model.language_model.layers.24.mlp.gate_proj": {
|
| 732 |
+
"count": 300941,
|
| 733 |
+
"input_features": 4096,
|
| 734 |
+
"sample_rows": 8
|
| 735 |
+
},
|
| 736 |
+
"model.language_model.layers.24.mlp.up_proj": {
|
| 737 |
+
"count": 300941,
|
| 738 |
+
"input_features": 4096,
|
| 739 |
+
"sample_rows": 8
|
| 740 |
+
},
|
| 741 |
+
"model.language_model.layers.25.linear_attn.in_proj_a": {
|
| 742 |
+
"count": 300941,
|
| 743 |
+
"input_features": 4096,
|
| 744 |
+
"sample_rows": 8
|
| 745 |
+
},
|
| 746 |
+
"model.language_model.layers.25.linear_attn.in_proj_b": {
|
| 747 |
+
"count": 300941,
|
| 748 |
+
"input_features": 4096,
|
| 749 |
+
"sample_rows": 8
|
| 750 |
+
},
|
| 751 |
+
"model.language_model.layers.25.linear_attn.in_proj_qkv": {
|
| 752 |
+
"count": 300941,
|
| 753 |
+
"input_features": 4096,
|
| 754 |
+
"sample_rows": 8
|
| 755 |
+
},
|
| 756 |
+
"model.language_model.layers.25.linear_attn.in_proj_z": {
|
| 757 |
+
"count": 300941,
|
| 758 |
+
"input_features": 4096,
|
| 759 |
+
"sample_rows": 8
|
| 760 |
+
},
|
| 761 |
+
"model.language_model.layers.25.linear_attn.out_proj": {
|
| 762 |
+
"count": 300941,
|
| 763 |
+
"input_features": 4096,
|
| 764 |
+
"sample_rows": 8
|
| 765 |
+
},
|
| 766 |
+
"model.language_model.layers.25.mlp.down_proj": {
|
| 767 |
+
"count": 300941,
|
| 768 |
+
"input_features": 12288,
|
| 769 |
+
"sample_rows": 8
|
| 770 |
+
},
|
| 771 |
+
"model.language_model.layers.25.mlp.gate_proj": {
|
| 772 |
+
"count": 300941,
|
| 773 |
+
"input_features": 4096,
|
| 774 |
+
"sample_rows": 8
|
| 775 |
+
},
|
| 776 |
+
"model.language_model.layers.25.mlp.up_proj": {
|
| 777 |
+
"count": 300941,
|
| 778 |
+
"input_features": 4096,
|
| 779 |
+
"sample_rows": 8
|
| 780 |
+
},
|
| 781 |
+
"model.language_model.layers.26.linear_attn.in_proj_a": {
|
| 782 |
+
"count": 300941,
|
| 783 |
+
"input_features": 4096,
|
| 784 |
+
"sample_rows": 8
|
| 785 |
+
},
|
| 786 |
+
"model.language_model.layers.26.linear_attn.in_proj_b": {
|
| 787 |
+
"count": 300941,
|
| 788 |
+
"input_features": 4096,
|
| 789 |
+
"sample_rows": 8
|
| 790 |
+
},
|
| 791 |
+
"model.language_model.layers.26.linear_attn.in_proj_qkv": {
|
| 792 |
+
"count": 300941,
|
| 793 |
+
"input_features": 4096,
|
| 794 |
+
"sample_rows": 8
|
| 795 |
+
},
|
| 796 |
+
"model.language_model.layers.26.linear_attn.in_proj_z": {
|
| 797 |
+
"count": 300941,
|
| 798 |
+
"input_features": 4096,
|
| 799 |
+
"sample_rows": 8
|
| 800 |
+
},
|
| 801 |
+
"model.language_model.layers.26.linear_attn.out_proj": {
|
| 802 |
+
"count": 300941,
|
| 803 |
+
"input_features": 4096,
|
| 804 |
+
"sample_rows": 8
|
| 805 |
+
},
|
| 806 |
+
"model.language_model.layers.26.mlp.down_proj": {
|
| 807 |
+
"count": 300941,
|
| 808 |
+
"input_features": 12288,
|
| 809 |
+
"sample_rows": 8
|
| 810 |
+
},
|
| 811 |
+
"model.language_model.layers.26.mlp.gate_proj": {
|
| 812 |
+
"count": 300941,
|
| 813 |
+
"input_features": 4096,
|
| 814 |
+
"sample_rows": 8
|
| 815 |
+
},
|
| 816 |
+
"model.language_model.layers.26.mlp.up_proj": {
|
| 817 |
+
"count": 300941,
|
| 818 |
+
"input_features": 4096,
|
| 819 |
+
"sample_rows": 8
|
| 820 |
+
},
|
| 821 |
+
"model.language_model.layers.27.mlp.down_proj": {
|
| 822 |
+
"count": 300941,
|
| 823 |
+
"input_features": 12288,
|
| 824 |
+
"sample_rows": 8
|
| 825 |
+
},
|
| 826 |
+
"model.language_model.layers.27.mlp.gate_proj": {
|
| 827 |
+
"count": 300941,
|
| 828 |
+
"input_features": 4096,
|
| 829 |
+
"sample_rows": 8
|
| 830 |
+
},
|
| 831 |
+
"model.language_model.layers.27.mlp.up_proj": {
|
| 832 |
+
"count": 300941,
|
| 833 |
+
"input_features": 4096,
|
| 834 |
+
"sample_rows": 8
|
| 835 |
+
},
|
| 836 |
+
"model.language_model.layers.27.self_attn.k_proj": {
|
| 837 |
+
"count": 300941,
|
| 838 |
+
"input_features": 4096,
|
| 839 |
+
"sample_rows": 8
|
| 840 |
+
},
|
| 841 |
+
"model.language_model.layers.27.self_attn.o_proj": {
|
| 842 |
+
"count": 300941,
|
| 843 |
+
"input_features": 4096,
|
| 844 |
+
"sample_rows": 8
|
| 845 |
+
},
|
| 846 |
+
"model.language_model.layers.27.self_attn.q_proj": {
|
| 847 |
+
"count": 300941,
|
| 848 |
+
"input_features": 4096,
|
| 849 |
+
"sample_rows": 8
|
| 850 |
+
},
|
| 851 |
+
"model.language_model.layers.27.self_attn.v_proj": {
|
| 852 |
+
"count": 300941,
|
| 853 |
+
"input_features": 4096,
|
| 854 |
+
"sample_rows": 8
|
| 855 |
+
},
|
| 856 |
+
"model.language_model.layers.28.linear_attn.in_proj_a": {
|
| 857 |
+
"count": 300941,
|
| 858 |
+
"input_features": 4096,
|
| 859 |
+
"sample_rows": 8
|
| 860 |
+
},
|
| 861 |
+
"model.language_model.layers.28.linear_attn.in_proj_b": {
|
| 862 |
+
"count": 300941,
|
| 863 |
+
"input_features": 4096,
|
| 864 |
+
"sample_rows": 8
|
| 865 |
+
},
|
| 866 |
+
"model.language_model.layers.28.linear_attn.in_proj_qkv": {
|
| 867 |
+
"count": 300941,
|
| 868 |
+
"input_features": 4096,
|
| 869 |
+
"sample_rows": 8
|
| 870 |
+
},
|
| 871 |
+
"model.language_model.layers.28.linear_attn.in_proj_z": {
|
| 872 |
+
"count": 300941,
|
| 873 |
+
"input_features": 4096,
|
| 874 |
+
"sample_rows": 8
|
| 875 |
+
},
|
| 876 |
+
"model.language_model.layers.28.linear_attn.out_proj": {
|
| 877 |
+
"count": 300941,
|
| 878 |
+
"input_features": 4096,
|
| 879 |
+
"sample_rows": 8
|
| 880 |
+
},
|
| 881 |
+
"model.language_model.layers.28.mlp.down_proj": {
|
| 882 |
+
"count": 300941,
|
| 883 |
+
"input_features": 12288,
|
| 884 |
+
"sample_rows": 8
|
| 885 |
+
},
|
| 886 |
+
"model.language_model.layers.28.mlp.gate_proj": {
|
| 887 |
+
"count": 300941,
|
| 888 |
+
"input_features": 4096,
|
| 889 |
+
"sample_rows": 8
|
| 890 |
+
},
|
| 891 |
+
"model.language_model.layers.28.mlp.up_proj": {
|
| 892 |
+
"count": 300941,
|
| 893 |
+
"input_features": 4096,
|
| 894 |
+
"sample_rows": 8
|
| 895 |
+
},
|
| 896 |
+
"model.language_model.layers.29.linear_attn.in_proj_a": {
|
| 897 |
+
"count": 300941,
|
| 898 |
+
"input_features": 4096,
|
| 899 |
+
"sample_rows": 8
|
| 900 |
+
},
|
| 901 |
+
"model.language_model.layers.29.linear_attn.in_proj_b": {
|
| 902 |
+
"count": 300941,
|
| 903 |
+
"input_features": 4096,
|
| 904 |
+
"sample_rows": 8
|
| 905 |
+
},
|
| 906 |
+
"model.language_model.layers.29.linear_attn.in_proj_qkv": {
|
| 907 |
+
"count": 300941,
|
| 908 |
+
"input_features": 4096,
|
| 909 |
+
"sample_rows": 8
|
| 910 |
+
},
|
| 911 |
+
"model.language_model.layers.29.linear_attn.in_proj_z": {
|
| 912 |
+
"count": 300941,
|
| 913 |
+
"input_features": 4096,
|
| 914 |
+
"sample_rows": 8
|
| 915 |
+
},
|
| 916 |
+
"model.language_model.layers.29.linear_attn.out_proj": {
|
| 917 |
+
"count": 300941,
|
| 918 |
+
"input_features": 4096,
|
| 919 |
+
"sample_rows": 8
|
| 920 |
+
},
|
| 921 |
+
"model.language_model.layers.29.mlp.down_proj": {
|
| 922 |
+
"count": 300941,
|
| 923 |
+
"input_features": 12288,
|
| 924 |
+
"sample_rows": 8
|
| 925 |
+
},
|
| 926 |
+
"model.language_model.layers.29.mlp.gate_proj": {
|
| 927 |
+
"count": 300941,
|
| 928 |
+
"input_features": 4096,
|
| 929 |
+
"sample_rows": 8
|
| 930 |
+
},
|
| 931 |
+
"model.language_model.layers.29.mlp.up_proj": {
|
| 932 |
+
"count": 300941,
|
| 933 |
+
"input_features": 4096,
|
| 934 |
+
"sample_rows": 8
|
| 935 |
+
},
|
| 936 |
+
"model.language_model.layers.3.mlp.down_proj": {
|
| 937 |
+
"count": 300941,
|
| 938 |
+
"input_features": 12288,
|
| 939 |
+
"sample_rows": 8
|
| 940 |
+
},
|
| 941 |
+
"model.language_model.layers.3.mlp.gate_proj": {
|
| 942 |
+
"count": 300941,
|
| 943 |
+
"input_features": 4096,
|
| 944 |
+
"sample_rows": 8
|
| 945 |
+
},
|
| 946 |
+
"model.language_model.layers.3.mlp.up_proj": {
|
| 947 |
+
"count": 300941,
|
| 948 |
+
"input_features": 4096,
|
| 949 |
+
"sample_rows": 8
|
| 950 |
+
},
|
| 951 |
+
"model.language_model.layers.3.self_attn.k_proj": {
|
| 952 |
+
"count": 300941,
|
| 953 |
+
"input_features": 4096,
|
| 954 |
+
"sample_rows": 8
|
| 955 |
+
},
|
| 956 |
+
"model.language_model.layers.3.self_attn.o_proj": {
|
| 957 |
+
"count": 300941,
|
| 958 |
+
"input_features": 4096,
|
| 959 |
+
"sample_rows": 8
|
| 960 |
+
},
|
| 961 |
+
"model.language_model.layers.3.self_attn.q_proj": {
|
| 962 |
+
"count": 300941,
|
| 963 |
+
"input_features": 4096,
|
| 964 |
+
"sample_rows": 8
|
| 965 |
+
},
|
| 966 |
+
"model.language_model.layers.3.self_attn.v_proj": {
|
| 967 |
+
"count": 300941,
|
| 968 |
+
"input_features": 4096,
|
| 969 |
+
"sample_rows": 8
|
| 970 |
+
},
|
| 971 |
+
"model.language_model.layers.30.linear_attn.in_proj_a": {
|
| 972 |
+
"count": 300941,
|
| 973 |
+
"input_features": 4096,
|
| 974 |
+
"sample_rows": 8
|
| 975 |
+
},
|
| 976 |
+
"model.language_model.layers.30.linear_attn.in_proj_b": {
|
| 977 |
+
"count": 300941,
|
| 978 |
+
"input_features": 4096,
|
| 979 |
+
"sample_rows": 8
|
| 980 |
+
},
|
| 981 |
+
"model.language_model.layers.30.linear_attn.in_proj_qkv": {
|
| 982 |
+
"count": 300941,
|
| 983 |
+
"input_features": 4096,
|
| 984 |
+
"sample_rows": 8
|
| 985 |
+
},
|
| 986 |
+
"model.language_model.layers.30.linear_attn.in_proj_z": {
|
| 987 |
+
"count": 300941,
|
| 988 |
+
"input_features": 4096,
|
| 989 |
+
"sample_rows": 8
|
| 990 |
+
},
|
| 991 |
+
"model.language_model.layers.30.linear_attn.out_proj": {
|
| 992 |
+
"count": 300941,
|
| 993 |
+
"input_features": 4096,
|
| 994 |
+
"sample_rows": 8
|
| 995 |
+
},
|
| 996 |
+
"model.language_model.layers.30.mlp.down_proj": {
|
| 997 |
+
"count": 300941,
|
| 998 |
+
"input_features": 12288,
|
| 999 |
+
"sample_rows": 8
|
| 1000 |
+
},
|
| 1001 |
+
"model.language_model.layers.30.mlp.gate_proj": {
|
| 1002 |
+
"count": 300941,
|
| 1003 |
+
"input_features": 4096,
|
| 1004 |
+
"sample_rows": 8
|
| 1005 |
+
},
|
| 1006 |
+
"model.language_model.layers.30.mlp.up_proj": {
|
| 1007 |
+
"count": 300941,
|
| 1008 |
+
"input_features": 4096,
|
| 1009 |
+
"sample_rows": 8
|
| 1010 |
+
},
|
| 1011 |
+
"model.language_model.layers.31.mlp.down_proj": {
|
| 1012 |
+
"count": 300941,
|
| 1013 |
+
"input_features": 12288,
|
| 1014 |
+
"sample_rows": 8
|
| 1015 |
+
},
|
| 1016 |
+
"model.language_model.layers.31.mlp.gate_proj": {
|
| 1017 |
+
"count": 300941,
|
| 1018 |
+
"input_features": 4096,
|
| 1019 |
+
"sample_rows": 8
|
| 1020 |
+
},
|
| 1021 |
+
"model.language_model.layers.31.mlp.up_proj": {
|
| 1022 |
+
"count": 300941,
|
| 1023 |
+
"input_features": 4096,
|
| 1024 |
+
"sample_rows": 8
|
| 1025 |
+
},
|
| 1026 |
+
"model.language_model.layers.31.self_attn.k_proj": {
|
| 1027 |
+
"count": 300941,
|
| 1028 |
+
"input_features": 4096,
|
| 1029 |
+
"sample_rows": 8
|
| 1030 |
+
},
|
| 1031 |
+
"model.language_model.layers.31.self_attn.o_proj": {
|
| 1032 |
+
"count": 300941,
|
| 1033 |
+
"input_features": 4096,
|
| 1034 |
+
"sample_rows": 8
|
| 1035 |
+
},
|
| 1036 |
+
"model.language_model.layers.31.self_attn.q_proj": {
|
| 1037 |
+
"count": 300941,
|
| 1038 |
+
"input_features": 4096,
|
| 1039 |
+
"sample_rows": 8
|
| 1040 |
+
},
|
| 1041 |
+
"model.language_model.layers.31.self_attn.v_proj": {
|
| 1042 |
+
"count": 300941,
|
| 1043 |
+
"input_features": 4096,
|
| 1044 |
+
"sample_rows": 8
|
| 1045 |
+
},
|
| 1046 |
+
"model.language_model.layers.4.linear_attn.in_proj_a": {
|
| 1047 |
+
"count": 300941,
|
| 1048 |
+
"input_features": 4096,
|
| 1049 |
+
"sample_rows": 8
|
| 1050 |
+
},
|
| 1051 |
+
"model.language_model.layers.4.linear_attn.in_proj_b": {
|
| 1052 |
+
"count": 300941,
|
| 1053 |
+
"input_features": 4096,
|
| 1054 |
+
"sample_rows": 8
|
| 1055 |
+
},
|
| 1056 |
+
"model.language_model.layers.4.linear_attn.in_proj_qkv": {
|
| 1057 |
+
"count": 300941,
|
| 1058 |
+
"input_features": 4096,
|
| 1059 |
+
"sample_rows": 8
|
| 1060 |
+
},
|
| 1061 |
+
"model.language_model.layers.4.linear_attn.in_proj_z": {
|
| 1062 |
+
"count": 300941,
|
| 1063 |
+
"input_features": 4096,
|
| 1064 |
+
"sample_rows": 8
|
| 1065 |
+
},
|
| 1066 |
+
"model.language_model.layers.4.linear_attn.out_proj": {
|
| 1067 |
+
"count": 300941,
|
| 1068 |
+
"input_features": 4096,
|
| 1069 |
+
"sample_rows": 8
|
| 1070 |
+
},
|
| 1071 |
+
"model.language_model.layers.4.mlp.down_proj": {
|
| 1072 |
+
"count": 300941,
|
| 1073 |
+
"input_features": 12288,
|
| 1074 |
+
"sample_rows": 8
|
| 1075 |
+
},
|
| 1076 |
+
"model.language_model.layers.4.mlp.gate_proj": {
|
| 1077 |
+
"count": 300941,
|
| 1078 |
+
"input_features": 4096,
|
| 1079 |
+
"sample_rows": 8
|
| 1080 |
+
},
|
| 1081 |
+
"model.language_model.layers.4.mlp.up_proj": {
|
| 1082 |
+
"count": 300941,
|
| 1083 |
+
"input_features": 4096,
|
| 1084 |
+
"sample_rows": 8
|
| 1085 |
+
},
|
| 1086 |
+
"model.language_model.layers.5.linear_attn.in_proj_a": {
|
| 1087 |
+
"count": 300941,
|
| 1088 |
+
"input_features": 4096,
|
| 1089 |
+
"sample_rows": 8
|
| 1090 |
+
},
|
| 1091 |
+
"model.language_model.layers.5.linear_attn.in_proj_b": {
|
| 1092 |
+
"count": 300941,
|
| 1093 |
+
"input_features": 4096,
|
| 1094 |
+
"sample_rows": 8
|
| 1095 |
+
},
|
| 1096 |
+
"model.language_model.layers.5.linear_attn.in_proj_qkv": {
|
| 1097 |
+
"count": 300941,
|
| 1098 |
+
"input_features": 4096,
|
| 1099 |
+
"sample_rows": 8
|
| 1100 |
+
},
|
| 1101 |
+
"model.language_model.layers.5.linear_attn.in_proj_z": {
|
| 1102 |
+
"count": 300941,
|
| 1103 |
+
"input_features": 4096,
|
| 1104 |
+
"sample_rows": 8
|
| 1105 |
+
},
|
| 1106 |
+
"model.language_model.layers.5.linear_attn.out_proj": {
|
| 1107 |
+
"count": 300941,
|
| 1108 |
+
"input_features": 4096,
|
| 1109 |
+
"sample_rows": 8
|
| 1110 |
+
},
|
| 1111 |
+
"model.language_model.layers.5.mlp.down_proj": {
|
| 1112 |
+
"count": 300941,
|
| 1113 |
+
"input_features": 12288,
|
| 1114 |
+
"sample_rows": 8
|
| 1115 |
+
},
|
| 1116 |
+
"model.language_model.layers.5.mlp.gate_proj": {
|
| 1117 |
+
"count": 300941,
|
| 1118 |
+
"input_features": 4096,
|
| 1119 |
+
"sample_rows": 8
|
| 1120 |
+
},
|
| 1121 |
+
"model.language_model.layers.5.mlp.up_proj": {
|
| 1122 |
+
"count": 300941,
|
| 1123 |
+
"input_features": 4096,
|
| 1124 |
+
"sample_rows": 8
|
| 1125 |
+
},
|
| 1126 |
+
"model.language_model.layers.6.linear_attn.in_proj_a": {
|
| 1127 |
+
"count": 300941,
|
| 1128 |
+
"input_features": 4096,
|
| 1129 |
+
"sample_rows": 8
|
| 1130 |
+
},
|
| 1131 |
+
"model.language_model.layers.6.linear_attn.in_proj_b": {
|
| 1132 |
+
"count": 300941,
|
| 1133 |
+
"input_features": 4096,
|
| 1134 |
+
"sample_rows": 8
|
| 1135 |
+
},
|
| 1136 |
+
"model.language_model.layers.6.linear_attn.in_proj_qkv": {
|
| 1137 |
+
"count": 300941,
|
| 1138 |
+
"input_features": 4096,
|
| 1139 |
+
"sample_rows": 8
|
| 1140 |
+
},
|
| 1141 |
+
"model.language_model.layers.6.linear_attn.in_proj_z": {
|
| 1142 |
+
"count": 300941,
|
| 1143 |
+
"input_features": 4096,
|
| 1144 |
+
"sample_rows": 8
|
| 1145 |
+
},
|
| 1146 |
+
"model.language_model.layers.6.linear_attn.out_proj": {
|
| 1147 |
+
"count": 300941,
|
| 1148 |
+
"input_features": 4096,
|
| 1149 |
+
"sample_rows": 8
|
| 1150 |
+
},
|
| 1151 |
+
"model.language_model.layers.6.mlp.down_proj": {
|
| 1152 |
+
"count": 300941,
|
| 1153 |
+
"input_features": 12288,
|
| 1154 |
+
"sample_rows": 8
|
| 1155 |
+
},
|
| 1156 |
+
"model.language_model.layers.6.mlp.gate_proj": {
|
| 1157 |
+
"count": 300941,
|
| 1158 |
+
"input_features": 4096,
|
| 1159 |
+
"sample_rows": 8
|
| 1160 |
+
},
|
| 1161 |
+
"model.language_model.layers.6.mlp.up_proj": {
|
| 1162 |
+
"count": 300941,
|
| 1163 |
+
"input_features": 4096,
|
| 1164 |
+
"sample_rows": 8
|
| 1165 |
+
},
|
| 1166 |
+
"model.language_model.layers.7.mlp.down_proj": {
|
| 1167 |
+
"count": 300941,
|
| 1168 |
+
"input_features": 12288,
|
| 1169 |
+
"sample_rows": 8
|
| 1170 |
+
},
|
| 1171 |
+
"model.language_model.layers.7.mlp.gate_proj": {
|
| 1172 |
+
"count": 300941,
|
| 1173 |
+
"input_features": 4096,
|
| 1174 |
+
"sample_rows": 8
|
| 1175 |
+
},
|
| 1176 |
+
"model.language_model.layers.7.mlp.up_proj": {
|
| 1177 |
+
"count": 300941,
|
| 1178 |
+
"input_features": 4096,
|
| 1179 |
+
"sample_rows": 8
|
| 1180 |
+
},
|
| 1181 |
+
"model.language_model.layers.7.self_attn.k_proj": {
|
| 1182 |
+
"count": 300941,
|
| 1183 |
+
"input_features": 4096,
|
| 1184 |
+
"sample_rows": 8
|
| 1185 |
+
},
|
| 1186 |
+
"model.language_model.layers.7.self_attn.o_proj": {
|
| 1187 |
+
"count": 300941,
|
| 1188 |
+
"input_features": 4096,
|
| 1189 |
+
"sample_rows": 8
|
| 1190 |
+
},
|
| 1191 |
+
"model.language_model.layers.7.self_attn.q_proj": {
|
| 1192 |
+
"count": 300941,
|
| 1193 |
+
"input_features": 4096,
|
| 1194 |
+
"sample_rows": 8
|
| 1195 |
+
},
|
| 1196 |
+
"model.language_model.layers.7.self_attn.v_proj": {
|
| 1197 |
+
"count": 300941,
|
| 1198 |
+
"input_features": 4096,
|
| 1199 |
+
"sample_rows": 8
|
| 1200 |
+
},
|
| 1201 |
+
"model.language_model.layers.8.linear_attn.in_proj_a": {
|
| 1202 |
+
"count": 300941,
|
| 1203 |
+
"input_features": 4096,
|
| 1204 |
+
"sample_rows": 8
|
| 1205 |
+
},
|
| 1206 |
+
"model.language_model.layers.8.linear_attn.in_proj_b": {
|
| 1207 |
+
"count": 300941,
|
| 1208 |
+
"input_features": 4096,
|
| 1209 |
+
"sample_rows": 8
|
| 1210 |
+
},
|
| 1211 |
+
"model.language_model.layers.8.linear_attn.in_proj_qkv": {
|
| 1212 |
+
"count": 300941,
|
| 1213 |
+
"input_features": 4096,
|
| 1214 |
+
"sample_rows": 8
|
| 1215 |
+
},
|
| 1216 |
+
"model.language_model.layers.8.linear_attn.in_proj_z": {
|
| 1217 |
+
"count": 300941,
|
| 1218 |
+
"input_features": 4096,
|
| 1219 |
+
"sample_rows": 8
|
| 1220 |
+
},
|
| 1221 |
+
"model.language_model.layers.8.linear_attn.out_proj": {
|
| 1222 |
+
"count": 300941,
|
| 1223 |
+
"input_features": 4096,
|
| 1224 |
+
"sample_rows": 8
|
| 1225 |
+
},
|
| 1226 |
+
"model.language_model.layers.8.mlp.down_proj": {
|
| 1227 |
+
"count": 300941,
|
| 1228 |
+
"input_features": 12288,
|
| 1229 |
+
"sample_rows": 8
|
| 1230 |
+
},
|
| 1231 |
+
"model.language_model.layers.8.mlp.gate_proj": {
|
| 1232 |
+
"count": 300941,
|
| 1233 |
+
"input_features": 4096,
|
| 1234 |
+
"sample_rows": 8
|
| 1235 |
+
},
|
| 1236 |
+
"model.language_model.layers.8.mlp.up_proj": {
|
| 1237 |
+
"count": 300941,
|
| 1238 |
+
"input_features": 4096,
|
| 1239 |
+
"sample_rows": 8
|
| 1240 |
+
},
|
| 1241 |
+
"model.language_model.layers.9.linear_attn.in_proj_a": {
|
| 1242 |
+
"count": 300941,
|
| 1243 |
+
"input_features": 4096,
|
| 1244 |
+
"sample_rows": 8
|
| 1245 |
+
},
|
| 1246 |
+
"model.language_model.layers.9.linear_attn.in_proj_b": {
|
| 1247 |
+
"count": 300941,
|
| 1248 |
+
"input_features": 4096,
|
| 1249 |
+
"sample_rows": 8
|
| 1250 |
+
},
|
| 1251 |
+
"model.language_model.layers.9.linear_attn.in_proj_qkv": {
|
| 1252 |
+
"count": 300941,
|
| 1253 |
+
"input_features": 4096,
|
| 1254 |
+
"sample_rows": 8
|
| 1255 |
+
},
|
| 1256 |
+
"model.language_model.layers.9.linear_attn.in_proj_z": {
|
| 1257 |
+
"count": 300941,
|
| 1258 |
+
"input_features": 4096,
|
| 1259 |
+
"sample_rows": 8
|
| 1260 |
+
},
|
| 1261 |
+
"model.language_model.layers.9.linear_attn.out_proj": {
|
| 1262 |
+
"count": 300941,
|
| 1263 |
+
"input_features": 4096,
|
| 1264 |
+
"sample_rows": 8
|
| 1265 |
+
},
|
| 1266 |
+
"model.language_model.layers.9.mlp.down_proj": {
|
| 1267 |
+
"count": 300941,
|
| 1268 |
+
"input_features": 12288,
|
| 1269 |
+
"sample_rows": 8
|
| 1270 |
+
},
|
| 1271 |
+
"model.language_model.layers.9.mlp.gate_proj": {
|
| 1272 |
+
"count": 300941,
|
| 1273 |
+
"input_features": 4096,
|
| 1274 |
+
"sample_rows": 8
|
| 1275 |
+
},
|
| 1276 |
+
"model.language_model.layers.9.mlp.up_proj": {
|
| 1277 |
+
"count": 300941,
|
| 1278 |
+
"input_features": 4096,
|
| 1279 |
+
"sample_rows": 8
|
| 1280 |
+
}
|
| 1281 |
+
},
|
| 1282 |
+
"native_fp32_tensors_restored": 48,
|
| 1283 |
+
"packages": {
|
| 1284 |
+
"accelerate": "1.7.0",
|
| 1285 |
+
"numpy": "1.26.4",
|
| 1286 |
+
"safetensors": "0.8.0",
|
| 1287 |
+
"torch": "2.11.0+cpu",
|
| 1288 |
+
"transformers": "5.12.1"
|
| 1289 |
+
},
|
| 1290 |
+
"parameter_elements_by_dtype": {
|
| 1291 |
+
"torch.bfloat16": 8953799424,
|
| 1292 |
+
"torch.float32": 3840
|
| 1293 |
+
},
|
| 1294 |
+
"reference_loader_sha256": "286b970571921b3a4ca3aa80a75a55c4191de404b4c95ad84828349bc504e046",
|
| 1295 |
+
"revision_evidence": {
|
| 1296 |
+
"chat_template.jinja": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 1297 |
+
"config.json": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 1298 |
+
"merges.txt": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 1299 |
+
"model.safetensors-00001-of-00004.safetensors": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 1300 |
+
"model.safetensors-00002-of-00004.safetensors": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 1301 |
+
"model.safetensors-00003-of-00004.safetensors": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 1302 |
+
"model.safetensors-00004-of-00004.safetensors": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 1303 |
+
"model.safetensors.index.json": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 1304 |
+
"tokenizer.json": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 1305 |
+
"tokenizer_config.json": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 1306 |
+
"vocab.json": "c202236235762e1c871ad0ccb60c8ee5ba337b9a"
|
| 1307 |
+
},
|
| 1308 |
+
"samples_sha256": "f5fa7dbda17bc7e8c09e6108eb648b49766e434e548786fe53746567049b67e4",
|
| 1309 |
+
"sampling": "Fixed approximately uniformly spaced token indices over the full deterministic record stream, not a first-record activation slice",
|
| 1310 |
+
"schema_version": 1,
|
| 1311 |
+
"script_sha256": "8e7b3e30bf5e74ed59d90b0b2e4e0a33a4985c7f7504fd7bd75d97a96087c5dc",
|
| 1312 |
+
"statistic": "float64 sum of FP32 squared linear-input activations over all full-record token positions; count is unpadded token rows. Mean diagonal second moment = sumsq / count.",
|
| 1313 |
+
"status": "complete",
|
| 1314 |
+
"text_tensor_count": 427,
|
| 1315 |
+
"tokenizer": {
|
| 1316 |
+
"applied_here": false,
|
| 1317 |
+
"file_sha256": {
|
| 1318 |
+
"chat_template.jinja": "a4aee8afcf2e0711942cf848899be66016f8d14a889ff9ede07bca099c28f715",
|
| 1319 |
+
"merges.txt": "a9d356d7bdf1ef4949e3e748e95b8e10ad9d4e2e838eddc38a0a7b6b94d1db8d",
|
| 1320 |
+
"tokenizer.json": "5f9e4d4901a92b997e463c1f46055088b6cca5ca61a6522d1b9f64c4bb81cb42",
|
| 1321 |
+
"tokenizer_config.json": "316230d6a809701f4db5ea8f8fc862bc3a6f3229c937c174e674ff3ca0a64ac8",
|
| 1322 |
+
"vocab.json": "ce99b4cb2983d118806ce0a8b777a35b093e2000a503ebde25853284c9dfa003"
|
| 1323 |
+
},
|
| 1324 |
+
"model_id": "Qwen/Qwen3.5-9B",
|
| 1325 |
+
"revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 1326 |
+
"vocabulary_order": "Original checkpoint lm_head row order, unchanged"
|
| 1327 |
+
},
|
| 1328 |
+
"total_input_tokens": 300941,
|
| 1329 |
+
"total_records": 614,
|
| 1330 |
+
"use": "TT precision sensitivity ranking; importance = sumsq/count. Not llama.cpp imatrix format; no training or quantized-weight optimization.",
|
| 1331 |
+
"weight_bytes_by_component": {
|
| 1332 |
+
"mtp": 486581248,
|
| 1333 |
+
"text": 17907614208,
|
| 1334 |
+
"vision": 912020960
|
| 1335 |
+
},
|
| 1336 |
+
"weight_integrity_note": "Revision sidecars plus config/index hashes; full multi-GB shard content hashes are not recomputed."
|
| 1337 |
+
}
|
checkpoint/provenance/importance.npz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ca51deb3c0f6b096bea1ff2312b1bab4a4a8fa083ba0e03b282ec4f8ef303779
|
| 3 |
+
size 10427338
|
checkpoint/provenance/original.safetensors.index.json
ADDED
|
@@ -0,0 +1,782 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_size": 19306216416
|
| 4 |
+
},
|
| 5 |
+
"weight_map": {
|
| 6 |
+
"lm_head.weight": "model.safetensors-00001-of-00004.safetensors",
|
| 7 |
+
"model.language_model.embed_tokens.weight": "model.safetensors-00001-of-00004.safetensors",
|
| 8 |
+
"model.language_model.layers.14.mlp.down_proj.weight": "model.safetensors-00001-of-00004.safetensors",
|
| 9 |
+
"model.language_model.layers.14.mlp.gate_proj.weight": "model.safetensors-00001-of-00004.safetensors",
|
| 10 |
+
"model.language_model.layers.14.mlp.up_proj.weight": "model.safetensors-00001-of-00004.safetensors",
|
| 11 |
+
"model.language_model.layers.15.mlp.down_proj.weight": "model.safetensors-00001-of-00004.safetensors",
|
| 12 |
+
"model.language_model.layers.15.mlp.gate_proj.weight": "model.safetensors-00001-of-00004.safetensors",
|
| 13 |
+
"model.language_model.layers.15.mlp.up_proj.weight": "model.safetensors-00001-of-00004.safetensors",
|
| 14 |
+
"model.language_model.layers.25.mlp.down_proj.weight": "model.safetensors-00001-of-00004.safetensors",
|
| 15 |
+
"model.language_model.layers.26.mlp.down_proj.weight": "model.safetensors-00001-of-00004.safetensors",
|
| 16 |
+
"model.language_model.layers.26.mlp.gate_proj.weight": "model.safetensors-00001-of-00004.safetensors",
|
| 17 |
+
"model.language_model.layers.26.mlp.up_proj.weight": "model.safetensors-00001-of-00004.safetensors",
|
| 18 |
+
"model.language_model.layers.27.mlp.down_proj.weight": "model.safetensors-00001-of-00004.safetensors",
|
| 19 |
+
"model.language_model.layers.27.mlp.gate_proj.weight": "model.safetensors-00001-of-00004.safetensors",
|
| 20 |
+
"model.language_model.layers.27.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 21 |
+
"model.language_model.layers.3.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 22 |
+
"model.language_model.layers.3.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 23 |
+
"model.language_model.layers.3.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 24 |
+
"model.language_model.layers.30.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 25 |
+
"model.language_model.layers.30.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 26 |
+
"model.language_model.layers.30.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 27 |
+
"model.language_model.layers.31.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 28 |
+
"model.language_model.layers.31.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 29 |
+
"model.language_model.layers.9.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 30 |
+
"model.language_model.layers.9.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 31 |
+
"model.language_model.layers.9.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 32 |
+
"mtp.layers.0.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 33 |
+
"mtp.layers.0.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 34 |
+
"mtp.layers.0.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 35 |
+
"model.language_model.layers.1.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 36 |
+
"model.language_model.layers.1.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 37 |
+
"model.language_model.layers.1.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 38 |
+
"model.language_model.layers.10.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 39 |
+
"model.language_model.layers.10.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 40 |
+
"model.language_model.layers.10.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 41 |
+
"model.language_model.layers.11.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 42 |
+
"model.language_model.layers.11.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 43 |
+
"model.language_model.layers.11.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 44 |
+
"model.language_model.layers.0.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 45 |
+
"model.language_model.layers.0.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 46 |
+
"model.language_model.layers.0.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 47 |
+
"model.language_model.layers.21.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 48 |
+
"model.language_model.layers.21.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 49 |
+
"model.language_model.layers.21.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 50 |
+
"model.language_model.layers.22.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 51 |
+
"model.language_model.layers.22.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 52 |
+
"model.language_model.layers.22.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 53 |
+
"model.language_model.layers.23.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 54 |
+
"model.language_model.layers.23.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 55 |
+
"model.language_model.layers.23.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 56 |
+
"model.language_model.layers.8.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 57 |
+
"model.language_model.layers.8.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 58 |
+
"model.language_model.layers.8.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 59 |
+
"model.language_model.layers.31.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 60 |
+
"model.language_model.layers.4.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 61 |
+
"model.language_model.layers.4.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 62 |
+
"model.language_model.layers.4.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 63 |
+
"model.language_model.layers.16.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 64 |
+
"model.language_model.layers.16.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 65 |
+
"model.language_model.layers.16.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 66 |
+
"model.language_model.layers.17.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 67 |
+
"model.language_model.layers.17.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 68 |
+
"model.language_model.layers.17.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 69 |
+
"model.language_model.layers.18.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 70 |
+
"model.language_model.layers.18.mlp.gate_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 71 |
+
"model.language_model.layers.18.mlp.up_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 72 |
+
"model.language_model.layers.19.mlp.down_proj.weight": "model.safetensors-00002-of-00004.safetensors",
|
| 73 |
+
"model.language_model.layers.19.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 74 |
+
"model.language_model.layers.19.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 75 |
+
"model.language_model.layers.24.mlp.down_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 76 |
+
"model.language_model.layers.24.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 77 |
+
"model.language_model.layers.24.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 78 |
+
"model.language_model.layers.25.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 79 |
+
"model.language_model.layers.25.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 80 |
+
"model.language_model.layers.5.mlp.down_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 81 |
+
"model.language_model.layers.5.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 82 |
+
"model.language_model.layers.5.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 83 |
+
"model.language_model.layers.6.mlp.down_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 84 |
+
"model.language_model.layers.6.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 85 |
+
"model.language_model.layers.6.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 86 |
+
"model.language_model.layers.7.mlp.down_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 87 |
+
"model.language_model.layers.7.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 88 |
+
"model.language_model.layers.7.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 89 |
+
"model.language_model.layers.2.mlp.down_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 90 |
+
"model.language_model.layers.2.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 91 |
+
"model.language_model.layers.2.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 92 |
+
"model.language_model.layers.20.mlp.down_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 93 |
+
"model.language_model.layers.20.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 94 |
+
"model.language_model.layers.20.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 95 |
+
"model.language_model.layers.28.mlp.down_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 96 |
+
"model.language_model.layers.28.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 97 |
+
"model.language_model.layers.28.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 98 |
+
"model.language_model.layers.29.mlp.down_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 99 |
+
"model.language_model.layers.29.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 100 |
+
"model.language_model.layers.29.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 101 |
+
"model.language_model.layers.12.mlp.down_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 102 |
+
"model.language_model.layers.12.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 103 |
+
"model.language_model.layers.12.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 104 |
+
"model.language_model.layers.13.mlp.down_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 105 |
+
"model.language_model.layers.13.mlp.gate_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 106 |
+
"model.language_model.layers.13.mlp.up_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 107 |
+
"model.language_model.layers.14.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 108 |
+
"model.language_model.layers.15.self_attn.q_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 109 |
+
"model.language_model.layers.26.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 110 |
+
"model.language_model.layers.27.self_attn.q_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 111 |
+
"model.language_model.layers.3.self_attn.q_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 112 |
+
"model.language_model.layers.30.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 113 |
+
"model.language_model.layers.9.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 114 |
+
"mtp.fc.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 115 |
+
"mtp.layers.0.self_attn.q_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 116 |
+
"model.language_model.layers.10.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 117 |
+
"model.language_model.layers.11.self_attn.q_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 118 |
+
"model.language_model.layers.0.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 119 |
+
"model.language_model.layers.1.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 120 |
+
"model.language_model.layers.22.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 121 |
+
"model.language_model.layers.23.self_attn.q_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 122 |
+
"model.language_model.layers.7.self_attn.q_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 123 |
+
"model.language_model.layers.8.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 124 |
+
"model.language_model.layers.31.self_attn.q_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 125 |
+
"model.language_model.layers.4.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 126 |
+
"model.language_model.layers.16.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 127 |
+
"model.language_model.layers.17.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 128 |
+
"model.language_model.layers.18.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 129 |
+
"model.language_model.layers.19.self_attn.q_proj.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 130 |
+
"model.language_model.layers.2.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 131 |
+
"model.language_model.layers.24.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 132 |
+
"model.language_model.layers.25.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 133 |
+
"model.language_model.layers.5.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 134 |
+
"model.language_model.layers.6.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 135 |
+
"model.language_model.layers.20.linear_attn.in_proj_qkv.weight": "model.safetensors-00003-of-00004.safetensors",
|
| 136 |
+
"model.language_model.layers.21.linear_attn.in_proj_qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 137 |
+
"model.language_model.layers.28.linear_attn.in_proj_qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 138 |
+
"model.language_model.layers.29.linear_attn.in_proj_qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 139 |
+
"model.language_model.layers.12.linear_attn.in_proj_qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 140 |
+
"model.language_model.layers.13.linear_attn.in_proj_qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 141 |
+
"model.visual.merger.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 142 |
+
"model.visual.merger.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 143 |
+
"model.language_model.layers.14.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 144 |
+
"model.language_model.layers.14.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 145 |
+
"model.language_model.layers.15.self_attn.o_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 146 |
+
"model.language_model.layers.26.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 147 |
+
"model.language_model.layers.26.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 148 |
+
"model.language_model.layers.27.self_attn.o_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 149 |
+
"model.language_model.layers.3.self_attn.o_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 150 |
+
"model.language_model.layers.30.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 151 |
+
"model.language_model.layers.30.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 152 |
+
"model.language_model.layers.9.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 153 |
+
"model.language_model.layers.9.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 154 |
+
"mtp.layers.0.self_attn.o_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 155 |
+
"model.language_model.layers.10.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 156 |
+
"model.language_model.layers.10.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 157 |
+
"model.language_model.layers.11.self_attn.o_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 158 |
+
"model.language_model.layers.0.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 159 |
+
"model.language_model.layers.0.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 160 |
+
"model.language_model.layers.1.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 161 |
+
"model.language_model.layers.1.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 162 |
+
"model.language_model.layers.21.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 163 |
+
"model.language_model.layers.22.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 164 |
+
"model.language_model.layers.22.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 165 |
+
"model.language_model.layers.23.self_attn.o_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 166 |
+
"model.language_model.layers.7.self_attn.o_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 167 |
+
"model.language_model.layers.8.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 168 |
+
"model.language_model.layers.8.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 169 |
+
"model.language_model.layers.31.self_attn.o_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 170 |
+
"model.language_model.layers.4.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 171 |
+
"model.language_model.layers.4.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 172 |
+
"model.language_model.layers.16.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 173 |
+
"model.language_model.layers.16.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 174 |
+
"model.language_model.layers.17.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 175 |
+
"model.language_model.layers.17.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 176 |
+
"model.language_model.layers.18.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 177 |
+
"model.language_model.layers.18.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 178 |
+
"model.language_model.layers.19.self_attn.o_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 179 |
+
"model.language_model.layers.2.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 180 |
+
"model.language_model.layers.24.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 181 |
+
"model.language_model.layers.24.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 182 |
+
"model.language_model.layers.25.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 183 |
+
"model.language_model.layers.25.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 184 |
+
"model.language_model.layers.5.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 185 |
+
"model.language_model.layers.5.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 186 |
+
"model.language_model.layers.6.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 187 |
+
"model.language_model.layers.6.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 188 |
+
"model.language_model.layers.2.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 189 |
+
"model.language_model.layers.20.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 190 |
+
"model.language_model.layers.20.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 191 |
+
"model.language_model.layers.21.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 192 |
+
"model.language_model.layers.28.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 193 |
+
"model.language_model.layers.28.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 194 |
+
"model.language_model.layers.29.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 195 |
+
"model.language_model.layers.29.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 196 |
+
"model.language_model.layers.12.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 197 |
+
"model.language_model.layers.12.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 198 |
+
"model.language_model.layers.13.linear_attn.in_proj_z.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 199 |
+
"model.language_model.layers.13.linear_attn.out_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 200 |
+
"model.visual.blocks.0.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 201 |
+
"model.visual.blocks.0.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 202 |
+
"model.visual.blocks.1.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 203 |
+
"model.visual.blocks.1.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 204 |
+
"model.visual.blocks.10.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 205 |
+
"model.visual.blocks.10.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 206 |
+
"model.visual.blocks.11.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 207 |
+
"model.visual.blocks.11.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 208 |
+
"model.visual.blocks.12.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 209 |
+
"model.visual.blocks.12.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 210 |
+
"model.visual.blocks.13.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 211 |
+
"model.visual.blocks.13.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 212 |
+
"model.visual.blocks.14.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 213 |
+
"model.visual.blocks.14.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 214 |
+
"model.visual.blocks.15.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 215 |
+
"model.visual.blocks.15.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 216 |
+
"model.visual.blocks.16.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 217 |
+
"model.visual.blocks.16.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 218 |
+
"model.visual.blocks.17.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 219 |
+
"model.visual.blocks.17.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 220 |
+
"model.visual.blocks.18.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 221 |
+
"model.visual.blocks.18.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 222 |
+
"model.visual.blocks.19.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 223 |
+
"model.visual.blocks.19.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 224 |
+
"model.visual.blocks.2.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 225 |
+
"model.visual.blocks.2.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 226 |
+
"model.visual.blocks.20.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 227 |
+
"model.visual.blocks.20.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 228 |
+
"model.visual.blocks.21.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 229 |
+
"model.visual.blocks.21.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 230 |
+
"model.visual.blocks.22.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 231 |
+
"model.visual.blocks.22.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 232 |
+
"model.visual.blocks.23.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 233 |
+
"model.visual.blocks.23.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 234 |
+
"model.visual.blocks.24.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 235 |
+
"model.visual.blocks.24.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 236 |
+
"model.visual.blocks.25.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 237 |
+
"model.visual.blocks.25.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 238 |
+
"model.visual.blocks.26.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 239 |
+
"model.visual.blocks.26.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 240 |
+
"model.visual.blocks.3.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 241 |
+
"model.visual.blocks.3.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 242 |
+
"model.visual.blocks.4.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 243 |
+
"model.visual.blocks.4.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 244 |
+
"model.visual.blocks.5.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 245 |
+
"model.visual.blocks.5.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 246 |
+
"model.visual.blocks.6.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 247 |
+
"model.visual.blocks.6.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 248 |
+
"model.visual.blocks.7.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 249 |
+
"model.visual.blocks.7.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 250 |
+
"model.visual.blocks.8.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 251 |
+
"model.visual.blocks.8.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 252 |
+
"model.visual.blocks.9.mlp.linear_fc1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 253 |
+
"model.visual.blocks.9.mlp.linear_fc2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 254 |
+
"model.language_model.layers.15.self_attn.k_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 255 |
+
"model.language_model.layers.15.self_attn.v_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 256 |
+
"model.language_model.layers.27.self_attn.k_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 257 |
+
"model.language_model.layers.27.self_attn.v_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 258 |
+
"model.language_model.layers.3.self_attn.k_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 259 |
+
"model.language_model.layers.3.self_attn.v_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 260 |
+
"mtp.layers.0.self_attn.k_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 261 |
+
"mtp.layers.0.self_attn.v_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 262 |
+
"model.language_model.layers.11.self_attn.k_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 263 |
+
"model.language_model.layers.11.self_attn.v_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 264 |
+
"model.language_model.layers.23.self_attn.k_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 265 |
+
"model.language_model.layers.23.self_attn.v_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 266 |
+
"model.language_model.layers.7.self_attn.k_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 267 |
+
"model.language_model.layers.7.self_attn.v_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 268 |
+
"model.language_model.layers.31.self_attn.k_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 269 |
+
"model.language_model.layers.31.self_attn.v_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 270 |
+
"model.language_model.layers.19.self_attn.k_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 271 |
+
"model.language_model.layers.19.self_attn.v_proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 272 |
+
"model.visual.blocks.0.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 273 |
+
"model.visual.blocks.1.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 274 |
+
"model.visual.blocks.10.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 275 |
+
"model.visual.blocks.11.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 276 |
+
"model.visual.blocks.12.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 277 |
+
"model.visual.blocks.13.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 278 |
+
"model.visual.blocks.14.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 279 |
+
"model.visual.blocks.15.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 280 |
+
"model.visual.blocks.16.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 281 |
+
"model.visual.blocks.17.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 282 |
+
"model.visual.blocks.18.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 283 |
+
"model.visual.blocks.19.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 284 |
+
"model.visual.blocks.2.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 285 |
+
"model.visual.blocks.20.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 286 |
+
"model.visual.blocks.21.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 287 |
+
"model.visual.blocks.22.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 288 |
+
"model.visual.blocks.23.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 289 |
+
"model.visual.blocks.24.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 290 |
+
"model.visual.blocks.25.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 291 |
+
"model.visual.blocks.26.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 292 |
+
"model.visual.blocks.3.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 293 |
+
"model.visual.blocks.4.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 294 |
+
"model.visual.blocks.5.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 295 |
+
"model.visual.blocks.6.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 296 |
+
"model.visual.blocks.7.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 297 |
+
"model.visual.blocks.8.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 298 |
+
"model.visual.blocks.9.attn.qkv.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 299 |
+
"model.visual.pos_embed.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 300 |
+
"model.visual.patch_embed.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 301 |
+
"model.visual.blocks.0.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 302 |
+
"model.visual.blocks.1.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 303 |
+
"model.visual.blocks.10.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 304 |
+
"model.visual.blocks.11.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 305 |
+
"model.visual.blocks.12.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 306 |
+
"model.visual.blocks.13.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 307 |
+
"model.visual.blocks.14.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 308 |
+
"model.visual.blocks.15.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 309 |
+
"model.visual.blocks.16.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 310 |
+
"model.visual.blocks.17.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 311 |
+
"model.visual.blocks.18.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 312 |
+
"model.visual.blocks.19.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 313 |
+
"model.visual.blocks.2.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 314 |
+
"model.visual.blocks.20.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 315 |
+
"model.visual.blocks.21.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 316 |
+
"model.visual.blocks.22.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 317 |
+
"model.visual.blocks.23.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 318 |
+
"model.visual.blocks.24.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 319 |
+
"model.visual.blocks.25.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 320 |
+
"model.visual.blocks.26.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 321 |
+
"model.visual.blocks.3.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 322 |
+
"model.visual.blocks.4.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 323 |
+
"model.visual.blocks.5.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 324 |
+
"model.visual.blocks.6.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 325 |
+
"model.visual.blocks.7.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 326 |
+
"model.visual.blocks.8.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 327 |
+
"model.visual.blocks.9.attn.proj.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 328 |
+
"model.language_model.layers.14.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 329 |
+
"model.language_model.layers.14.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 330 |
+
"model.language_model.layers.26.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 331 |
+
"model.language_model.layers.26.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 332 |
+
"model.language_model.layers.30.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 333 |
+
"model.language_model.layers.30.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 334 |
+
"model.language_model.layers.10.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 335 |
+
"model.language_model.layers.10.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 336 |
+
"model.language_model.layers.0.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 337 |
+
"model.language_model.layers.0.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 338 |
+
"model.language_model.layers.1.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 339 |
+
"model.language_model.layers.1.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 340 |
+
"model.language_model.layers.22.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 341 |
+
"model.language_model.layers.22.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 342 |
+
"model.language_model.layers.8.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 343 |
+
"model.language_model.layers.8.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 344 |
+
"model.language_model.layers.9.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 345 |
+
"model.language_model.layers.9.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 346 |
+
"model.language_model.layers.4.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 347 |
+
"model.language_model.layers.4.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 348 |
+
"model.language_model.layers.16.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 349 |
+
"model.language_model.layers.16.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 350 |
+
"model.language_model.layers.17.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 351 |
+
"model.language_model.layers.17.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 352 |
+
"model.language_model.layers.18.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 353 |
+
"model.language_model.layers.18.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 354 |
+
"model.language_model.layers.2.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 355 |
+
"model.language_model.layers.2.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 356 |
+
"model.language_model.layers.24.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 357 |
+
"model.language_model.layers.24.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 358 |
+
"model.language_model.layers.25.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 359 |
+
"model.language_model.layers.25.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 360 |
+
"model.language_model.layers.5.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 361 |
+
"model.language_model.layers.5.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 362 |
+
"model.language_model.layers.6.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 363 |
+
"model.language_model.layers.6.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 364 |
+
"model.language_model.layers.20.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 365 |
+
"model.language_model.layers.20.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 366 |
+
"model.language_model.layers.21.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 367 |
+
"model.language_model.layers.21.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 368 |
+
"model.language_model.layers.28.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 369 |
+
"model.language_model.layers.28.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 370 |
+
"model.language_model.layers.29.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 371 |
+
"model.language_model.layers.29.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 372 |
+
"model.language_model.layers.12.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 373 |
+
"model.language_model.layers.12.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 374 |
+
"model.language_model.layers.13.linear_attn.in_proj_b.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 375 |
+
"model.language_model.layers.13.linear_attn.in_proj_a.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 376 |
+
"model.language_model.layers.14.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 377 |
+
"model.language_model.layers.16.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 378 |
+
"model.language_model.layers.26.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 379 |
+
"model.language_model.layers.30.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 380 |
+
"model.language_model.layers.10.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 381 |
+
"model.language_model.layers.0.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 382 |
+
"model.language_model.layers.1.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 383 |
+
"model.language_model.layers.22.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 384 |
+
"model.language_model.layers.8.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 385 |
+
"model.language_model.layers.9.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 386 |
+
"model.language_model.layers.4.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 387 |
+
"model.language_model.layers.5.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 388 |
+
"model.language_model.layers.17.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 389 |
+
"model.language_model.layers.18.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 390 |
+
"model.language_model.layers.2.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 391 |
+
"model.language_model.layers.24.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 392 |
+
"model.language_model.layers.25.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 393 |
+
"model.language_model.layers.6.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 394 |
+
"model.language_model.layers.20.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 395 |
+
"model.language_model.layers.21.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 396 |
+
"model.language_model.layers.28.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 397 |
+
"model.language_model.layers.29.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 398 |
+
"model.language_model.layers.12.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 399 |
+
"model.language_model.layers.13.linear_attn.conv1d.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 400 |
+
"model.visual.merger.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 401 |
+
"model.visual.blocks.0.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 402 |
+
"model.visual.blocks.1.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 403 |
+
"model.visual.blocks.10.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 404 |
+
"model.visual.blocks.11.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 405 |
+
"model.visual.blocks.12.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 406 |
+
"model.visual.blocks.13.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 407 |
+
"model.visual.blocks.14.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 408 |
+
"model.visual.blocks.15.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 409 |
+
"model.visual.blocks.16.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 410 |
+
"model.visual.blocks.17.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 411 |
+
"model.visual.blocks.18.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 412 |
+
"model.visual.blocks.19.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 413 |
+
"model.visual.blocks.2.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 414 |
+
"model.visual.blocks.20.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 415 |
+
"model.visual.blocks.21.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 416 |
+
"model.visual.blocks.22.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 417 |
+
"model.visual.blocks.23.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 418 |
+
"model.visual.blocks.24.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 419 |
+
"model.visual.blocks.25.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 420 |
+
"model.visual.blocks.26.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 421 |
+
"model.visual.blocks.3.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 422 |
+
"model.visual.blocks.4.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 423 |
+
"model.visual.blocks.5.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 424 |
+
"model.visual.blocks.6.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 425 |
+
"model.visual.blocks.7.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 426 |
+
"model.visual.blocks.8.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 427 |
+
"model.visual.blocks.9.mlp.linear_fc1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 428 |
+
"model.language_model.layers.14.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 429 |
+
"model.language_model.layers.14.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 430 |
+
"model.language_model.layers.15.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 431 |
+
"model.language_model.layers.15.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 432 |
+
"model.language_model.layers.16.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 433 |
+
"model.language_model.layers.26.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 434 |
+
"model.language_model.layers.26.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 435 |
+
"model.language_model.layers.27.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 436 |
+
"model.language_model.layers.27.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 437 |
+
"model.language_model.layers.3.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 438 |
+
"model.language_model.layers.3.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 439 |
+
"model.language_model.layers.30.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 440 |
+
"model.language_model.layers.30.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 441 |
+
"model.language_model.layers.31.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 442 |
+
"model.language_model.layers.9.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 443 |
+
"mtp.layers.0.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 444 |
+
"mtp.layers.0.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 445 |
+
"mtp.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 446 |
+
"mtp.pre_fc_norm_embedding.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 447 |
+
"mtp.pre_fc_norm_hidden.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 448 |
+
"model.language_model.layers.10.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 449 |
+
"model.language_model.layers.10.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 450 |
+
"model.language_model.layers.11.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 451 |
+
"model.language_model.layers.11.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 452 |
+
"model.language_model.layers.0.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 453 |
+
"model.language_model.layers.0.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 454 |
+
"model.language_model.layers.1.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 455 |
+
"model.language_model.layers.1.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 456 |
+
"model.language_model.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 457 |
+
"model.visual.merger.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 458 |
+
"model.language_model.layers.21.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 459 |
+
"model.language_model.layers.22.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 460 |
+
"model.language_model.layers.22.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 461 |
+
"model.language_model.layers.23.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 462 |
+
"model.language_model.layers.23.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 463 |
+
"model.language_model.layers.7.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 464 |
+
"model.language_model.layers.8.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 465 |
+
"model.language_model.layers.8.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 466 |
+
"model.language_model.layers.9.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 467 |
+
"model.language_model.layers.31.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 468 |
+
"model.language_model.layers.4.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 469 |
+
"model.language_model.layers.4.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 470 |
+
"model.language_model.layers.5.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 471 |
+
"model.language_model.layers.16.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 472 |
+
"model.language_model.layers.17.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 473 |
+
"model.language_model.layers.17.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 474 |
+
"model.language_model.layers.18.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 475 |
+
"model.language_model.layers.18.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 476 |
+
"model.language_model.layers.19.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 477 |
+
"model.language_model.layers.19.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 478 |
+
"model.language_model.layers.2.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 479 |
+
"model.language_model.layers.24.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 480 |
+
"model.language_model.layers.24.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 481 |
+
"model.language_model.layers.25.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 482 |
+
"model.language_model.layers.25.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 483 |
+
"model.language_model.layers.5.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 484 |
+
"model.language_model.layers.6.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 485 |
+
"model.language_model.layers.6.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 486 |
+
"model.language_model.layers.7.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 487 |
+
"model.language_model.layers.2.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 488 |
+
"model.language_model.layers.20.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 489 |
+
"model.language_model.layers.20.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 490 |
+
"model.language_model.layers.21.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 491 |
+
"model.language_model.layers.28.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 492 |
+
"model.language_model.layers.28.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 493 |
+
"model.language_model.layers.29.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 494 |
+
"model.language_model.layers.29.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 495 |
+
"model.language_model.layers.12.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 496 |
+
"model.language_model.layers.12.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 497 |
+
"model.language_model.layers.13.input_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 498 |
+
"model.language_model.layers.13.post_attention_layernorm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 499 |
+
"model.visual.blocks.0.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 500 |
+
"model.visual.blocks.1.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 501 |
+
"model.visual.blocks.10.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 502 |
+
"model.visual.blocks.11.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 503 |
+
"model.visual.blocks.12.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 504 |
+
"model.visual.blocks.13.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 505 |
+
"model.visual.blocks.14.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 506 |
+
"model.visual.blocks.15.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 507 |
+
"model.visual.blocks.16.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 508 |
+
"model.visual.blocks.17.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 509 |
+
"model.visual.blocks.18.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 510 |
+
"model.visual.blocks.19.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 511 |
+
"model.visual.blocks.2.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 512 |
+
"model.visual.blocks.20.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 513 |
+
"model.visual.blocks.21.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 514 |
+
"model.visual.blocks.22.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 515 |
+
"model.visual.blocks.23.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 516 |
+
"model.visual.blocks.24.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 517 |
+
"model.visual.blocks.25.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 518 |
+
"model.visual.blocks.26.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 519 |
+
"model.visual.blocks.3.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 520 |
+
"model.visual.blocks.4.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 521 |
+
"model.visual.blocks.5.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 522 |
+
"model.visual.blocks.6.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 523 |
+
"model.visual.blocks.7.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 524 |
+
"model.visual.blocks.8.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 525 |
+
"model.visual.blocks.9.attn.qkv.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 526 |
+
"model.visual.blocks.0.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 527 |
+
"model.visual.blocks.0.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 528 |
+
"model.visual.blocks.0.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 529 |
+
"model.visual.blocks.0.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 530 |
+
"model.visual.blocks.0.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 531 |
+
"model.visual.blocks.0.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 532 |
+
"model.visual.blocks.1.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 533 |
+
"model.visual.blocks.1.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 534 |
+
"model.visual.blocks.1.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 535 |
+
"model.visual.blocks.1.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 536 |
+
"model.visual.blocks.1.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 537 |
+
"model.visual.blocks.1.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 538 |
+
"model.visual.blocks.10.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 539 |
+
"model.visual.blocks.10.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 540 |
+
"model.visual.blocks.10.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 541 |
+
"model.visual.blocks.10.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 542 |
+
"model.visual.blocks.10.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 543 |
+
"model.visual.blocks.10.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 544 |
+
"model.visual.blocks.11.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 545 |
+
"model.visual.blocks.11.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 546 |
+
"model.visual.blocks.11.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 547 |
+
"model.visual.blocks.11.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 548 |
+
"model.visual.blocks.11.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 549 |
+
"model.visual.blocks.11.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 550 |
+
"model.visual.blocks.12.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 551 |
+
"model.visual.blocks.12.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 552 |
+
"model.visual.blocks.12.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 553 |
+
"model.visual.blocks.12.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 554 |
+
"model.visual.blocks.12.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 555 |
+
"model.visual.blocks.12.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 556 |
+
"model.visual.blocks.13.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 557 |
+
"model.visual.blocks.13.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 558 |
+
"model.visual.blocks.13.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 559 |
+
"model.visual.blocks.13.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 560 |
+
"model.visual.blocks.13.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 561 |
+
"model.visual.blocks.13.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 562 |
+
"model.visual.blocks.14.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 563 |
+
"model.visual.blocks.14.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 564 |
+
"model.visual.blocks.14.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 565 |
+
"model.visual.blocks.14.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 566 |
+
"model.visual.blocks.14.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 567 |
+
"model.visual.blocks.14.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 568 |
+
"model.visual.blocks.15.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 569 |
+
"model.visual.blocks.15.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 570 |
+
"model.visual.blocks.15.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 571 |
+
"model.visual.blocks.15.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 572 |
+
"model.visual.blocks.15.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 573 |
+
"model.visual.blocks.15.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 574 |
+
"model.visual.blocks.16.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 575 |
+
"model.visual.blocks.16.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 576 |
+
"model.visual.blocks.16.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 577 |
+
"model.visual.blocks.16.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 578 |
+
"model.visual.blocks.16.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 579 |
+
"model.visual.blocks.16.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 580 |
+
"model.visual.blocks.17.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 581 |
+
"model.visual.blocks.17.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 582 |
+
"model.visual.blocks.17.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 583 |
+
"model.visual.blocks.17.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 584 |
+
"model.visual.blocks.17.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 585 |
+
"model.visual.blocks.17.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 586 |
+
"model.visual.blocks.18.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 587 |
+
"model.visual.blocks.18.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 588 |
+
"model.visual.blocks.18.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 589 |
+
"model.visual.blocks.18.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 590 |
+
"model.visual.blocks.18.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 591 |
+
"model.visual.blocks.18.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 592 |
+
"model.visual.blocks.19.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 593 |
+
"model.visual.blocks.19.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 594 |
+
"model.visual.blocks.19.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 595 |
+
"model.visual.blocks.19.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 596 |
+
"model.visual.blocks.19.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 597 |
+
"model.visual.blocks.19.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 598 |
+
"model.visual.blocks.2.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 599 |
+
"model.visual.blocks.2.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 600 |
+
"model.visual.blocks.2.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 601 |
+
"model.visual.blocks.2.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 602 |
+
"model.visual.blocks.2.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 603 |
+
"model.visual.blocks.2.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 604 |
+
"model.visual.blocks.20.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 605 |
+
"model.visual.blocks.20.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 606 |
+
"model.visual.blocks.20.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 607 |
+
"model.visual.blocks.20.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 608 |
+
"model.visual.blocks.20.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 609 |
+
"model.visual.blocks.20.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 610 |
+
"model.visual.blocks.21.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 611 |
+
"model.visual.blocks.21.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 612 |
+
"model.visual.blocks.21.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 613 |
+
"model.visual.blocks.21.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 614 |
+
"model.visual.blocks.21.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 615 |
+
"model.visual.blocks.21.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 616 |
+
"model.visual.blocks.22.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 617 |
+
"model.visual.blocks.22.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 618 |
+
"model.visual.blocks.22.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 619 |
+
"model.visual.blocks.22.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 620 |
+
"model.visual.blocks.22.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 621 |
+
"model.visual.blocks.22.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 622 |
+
"model.visual.blocks.23.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 623 |
+
"model.visual.blocks.23.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 624 |
+
"model.visual.blocks.23.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 625 |
+
"model.visual.blocks.23.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 626 |
+
"model.visual.blocks.23.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 627 |
+
"model.visual.blocks.23.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 628 |
+
"model.visual.blocks.24.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 629 |
+
"model.visual.blocks.24.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 630 |
+
"model.visual.blocks.24.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 631 |
+
"model.visual.blocks.24.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 632 |
+
"model.visual.blocks.24.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 633 |
+
"model.visual.blocks.24.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 634 |
+
"model.visual.blocks.25.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 635 |
+
"model.visual.blocks.25.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 636 |
+
"model.visual.blocks.25.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 637 |
+
"model.visual.blocks.25.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 638 |
+
"model.visual.blocks.25.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 639 |
+
"model.visual.blocks.25.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 640 |
+
"model.visual.blocks.26.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 641 |
+
"model.visual.blocks.26.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 642 |
+
"model.visual.blocks.26.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 643 |
+
"model.visual.blocks.26.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 644 |
+
"model.visual.blocks.26.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 645 |
+
"model.visual.blocks.26.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 646 |
+
"model.visual.blocks.3.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 647 |
+
"model.visual.blocks.3.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 648 |
+
"model.visual.blocks.3.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 649 |
+
"model.visual.blocks.3.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 650 |
+
"model.visual.blocks.3.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 651 |
+
"model.visual.blocks.3.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 652 |
+
"model.visual.blocks.4.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 653 |
+
"model.visual.blocks.4.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 654 |
+
"model.visual.blocks.4.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 655 |
+
"model.visual.blocks.4.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 656 |
+
"model.visual.blocks.4.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 657 |
+
"model.visual.blocks.4.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 658 |
+
"model.visual.blocks.5.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 659 |
+
"model.visual.blocks.5.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 660 |
+
"model.visual.blocks.5.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 661 |
+
"model.visual.blocks.5.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 662 |
+
"model.visual.blocks.5.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 663 |
+
"model.visual.blocks.5.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 664 |
+
"model.visual.blocks.6.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 665 |
+
"model.visual.blocks.6.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 666 |
+
"model.visual.blocks.6.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 667 |
+
"model.visual.blocks.6.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 668 |
+
"model.visual.blocks.6.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 669 |
+
"model.visual.blocks.6.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 670 |
+
"model.visual.blocks.7.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 671 |
+
"model.visual.blocks.7.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 672 |
+
"model.visual.blocks.7.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 673 |
+
"model.visual.blocks.7.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 674 |
+
"model.visual.blocks.7.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 675 |
+
"model.visual.blocks.7.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 676 |
+
"model.visual.blocks.8.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 677 |
+
"model.visual.blocks.8.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 678 |
+
"model.visual.blocks.8.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 679 |
+
"model.visual.blocks.8.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 680 |
+
"model.visual.blocks.8.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 681 |
+
"model.visual.blocks.8.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 682 |
+
"model.visual.blocks.9.attn.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 683 |
+
"model.visual.blocks.9.mlp.linear_fc2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 684 |
+
"model.visual.blocks.9.norm1.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 685 |
+
"model.visual.blocks.9.norm1.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 686 |
+
"model.visual.blocks.9.norm2.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 687 |
+
"model.visual.blocks.9.norm2.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 688 |
+
"model.visual.merger.norm.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 689 |
+
"model.visual.merger.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 690 |
+
"model.visual.patch_embed.proj.bias": "model.safetensors-00004-of-00004.safetensors",
|
| 691 |
+
"model.language_model.layers.14.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 692 |
+
"model.language_model.layers.15.self_attn.k_norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 693 |
+
"model.language_model.layers.15.self_attn.q_norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 694 |
+
"model.language_model.layers.26.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 695 |
+
"model.language_model.layers.27.self_attn.k_norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 696 |
+
"model.language_model.layers.27.self_attn.q_norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 697 |
+
"model.language_model.layers.3.self_attn.k_norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 698 |
+
"model.language_model.layers.3.self_attn.q_norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 699 |
+
"model.language_model.layers.30.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 700 |
+
"model.language_model.layers.9.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 701 |
+
"mtp.layers.0.self_attn.k_norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 702 |
+
"mtp.layers.0.self_attn.q_norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 703 |
+
"model.language_model.layers.10.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 704 |
+
"model.language_model.layers.11.self_attn.k_norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 705 |
+
"model.language_model.layers.11.self_attn.q_norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 706 |
+
"model.language_model.layers.0.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 707 |
+
"model.language_model.layers.1.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 708 |
+
"model.language_model.layers.22.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 709 |
+
"model.language_model.layers.23.self_attn.k_norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 710 |
+
"model.language_model.layers.7.self_attn.k_norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 711 |
+
"model.language_model.layers.7.self_attn.q_norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 712 |
+
"model.language_model.layers.8.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 713 |
+
"model.language_model.layers.31.self_attn.k_norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 714 |
+
"model.language_model.layers.31.self_attn.q_norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 715 |
+
"model.language_model.layers.4.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 716 |
+
"model.language_model.layers.16.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 717 |
+
"model.language_model.layers.17.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 718 |
+
"model.language_model.layers.18.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 719 |
+
"model.language_model.layers.19.self_attn.k_norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 720 |
+
"model.language_model.layers.19.self_attn.q_norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 721 |
+
"model.language_model.layers.2.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 722 |
+
"model.language_model.layers.23.self_attn.q_norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 723 |
+
"model.language_model.layers.24.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 724 |
+
"model.language_model.layers.25.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 725 |
+
"model.language_model.layers.5.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 726 |
+
"model.language_model.layers.6.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 727 |
+
"model.language_model.layers.20.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 728 |
+
"model.language_model.layers.21.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 729 |
+
"model.language_model.layers.28.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 730 |
+
"model.language_model.layers.29.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 731 |
+
"model.language_model.layers.12.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 732 |
+
"model.language_model.layers.13.linear_attn.norm.weight": "model.safetensors-00004-of-00004.safetensors",
|
| 733 |
+
"model.language_model.layers.16.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 734 |
+
"model.language_model.layers.26.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 735 |
+
"model.language_model.layers.28.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 736 |
+
"model.language_model.layers.30.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 737 |
+
"model.language_model.layers.10.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 738 |
+
"model.language_model.layers.12.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 739 |
+
"model.language_model.layers.0.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 740 |
+
"model.language_model.layers.1.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 741 |
+
"model.language_model.layers.22.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 742 |
+
"model.language_model.layers.8.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 743 |
+
"model.language_model.layers.9.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 744 |
+
"model.language_model.layers.4.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 745 |
+
"model.language_model.layers.5.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 746 |
+
"model.language_model.layers.17.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 747 |
+
"model.language_model.layers.18.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 748 |
+
"model.language_model.layers.2.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 749 |
+
"model.language_model.layers.24.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 750 |
+
"model.language_model.layers.25.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 751 |
+
"model.language_model.layers.6.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 752 |
+
"model.language_model.layers.20.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 753 |
+
"model.language_model.layers.21.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 754 |
+
"model.language_model.layers.29.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 755 |
+
"model.language_model.layers.13.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 756 |
+
"model.language_model.layers.14.linear_attn.A_log": "model.safetensors-00004-of-00004.safetensors",
|
| 757 |
+
"model.language_model.layers.14.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
|
| 758 |
+
"model.language_model.layers.16.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
|
| 759 |
+
"model.language_model.layers.26.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
|
| 760 |
+
"model.language_model.layers.30.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
|
| 761 |
+
"model.language_model.layers.10.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
|
| 762 |
+
"model.language_model.layers.0.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
|
| 763 |
+
"model.language_model.layers.1.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
|
| 764 |
+
"model.language_model.layers.22.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
|
| 765 |
+
"model.language_model.layers.8.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
|
| 766 |
+
"model.language_model.layers.9.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
|
| 767 |
+
"model.language_model.layers.4.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
|
| 768 |
+
"model.language_model.layers.5.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
|
| 769 |
+
"model.language_model.layers.17.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
|
| 770 |
+
"model.language_model.layers.18.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
|
| 771 |
+
"model.language_model.layers.2.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
|
| 772 |
+
"model.language_model.layers.24.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
|
| 773 |
+
"model.language_model.layers.25.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
|
| 774 |
+
"model.language_model.layers.6.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
|
| 775 |
+
"model.language_model.layers.20.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
|
| 776 |
+
"model.language_model.layers.21.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
|
| 777 |
+
"model.language_model.layers.28.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
|
| 778 |
+
"model.language_model.layers.29.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
|
| 779 |
+
"model.language_model.layers.12.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors",
|
| 780 |
+
"model.language_model.layers.13.linear_attn.dt_bias": "model.safetensors-00004-of-00004.safetensors"
|
| 781 |
+
}
|
| 782 |
+
}
|
checkpoint/provenance/overrides.json
ADDED
|
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"families": {
|
| 3 |
+
"attention": "bfp8",
|
| 4 |
+
"gdn": "bfp8",
|
| 5 |
+
"gdn_output": "bfp8",
|
| 6 |
+
"lm_head": "bfp8"
|
| 7 |
+
},
|
| 8 |
+
"layers": {
|
| 9 |
+
"31": {
|
| 10 |
+
"mlp_down": "bfp8",
|
| 11 |
+
"mlp_gate_up": "bfp8"
|
| 12 |
+
},
|
| 13 |
+
"18": {
|
| 14 |
+
"mlp_gate_up": "bfp8",
|
| 15 |
+
"mlp_down": "bfp8"
|
| 16 |
+
},
|
| 17 |
+
"17": {
|
| 18 |
+
"mlp_gate_up": "bfp8"
|
| 19 |
+
},
|
| 20 |
+
"19": {
|
| 21 |
+
"mlp_gate_up": "bfp8",
|
| 22 |
+
"mlp_down": "bfp8"
|
| 23 |
+
},
|
| 24 |
+
"22": {
|
| 25 |
+
"mlp_gate_up": "bfp8",
|
| 26 |
+
"mlp_down": "bfp8"
|
| 27 |
+
},
|
| 28 |
+
"16": {
|
| 29 |
+
"mlp_gate_up": "bfp8",
|
| 30 |
+
"mlp_down": "bfp8"
|
| 31 |
+
},
|
| 32 |
+
"6": {
|
| 33 |
+
"mlp_down": "bfp8"
|
| 34 |
+
},
|
| 35 |
+
"20": {
|
| 36 |
+
"mlp_gate_up": "bfp8"
|
| 37 |
+
},
|
| 38 |
+
"15": {
|
| 39 |
+
"mlp_gate_up": "bfp8"
|
| 40 |
+
},
|
| 41 |
+
"21": {
|
| 42 |
+
"mlp_gate_up": "bfp8"
|
| 43 |
+
},
|
| 44 |
+
"14": {
|
| 45 |
+
"mlp_gate_up": "bfp8"
|
| 46 |
+
},
|
| 47 |
+
"30": {
|
| 48 |
+
"mlp_gate_up": "bfp8",
|
| 49 |
+
"mlp_down": "bfp8"
|
| 50 |
+
},
|
| 51 |
+
"13": {
|
| 52 |
+
"mlp_gate_up": "bfp8"
|
| 53 |
+
},
|
| 54 |
+
"12": {
|
| 55 |
+
"mlp_gate_up": "bfp8"
|
| 56 |
+
},
|
| 57 |
+
"29": {
|
| 58 |
+
"mlp_gate_up": "bfp8"
|
| 59 |
+
},
|
| 60 |
+
"23": {
|
| 61 |
+
"mlp_gate_up": "bfp8"
|
| 62 |
+
},
|
| 63 |
+
"28": {
|
| 64 |
+
"mlp_gate_up": "bfp8"
|
| 65 |
+
},
|
| 66 |
+
"27": {
|
| 67 |
+
"mlp_gate_up": "bfp8"
|
| 68 |
+
},
|
| 69 |
+
"1": {
|
| 70 |
+
"mlp_down": "bfp8"
|
| 71 |
+
}
|
| 72 |
+
}
|
| 73 |
+
}
|
checkpoint/provenance/precision-plan.json
ADDED
|
@@ -0,0 +1,233 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"schema_version": 1,
|
| 3 |
+
"status": "implemented_not_device_validated",
|
| 4 |
+
"scope": "Qwen3.5-9B text-only, one Blackhole P150, no MTP, no vision, no upload",
|
| 5 |
+
"upstream_revision": "c202236235762e1c871ad0ccb60c8ee5ba337b9a",
|
| 6 |
+
"weights_host": "/var/tmp/qwen9b-hf-publish/clean-cache/weights/Qwen3.5-9B",
|
| 7 |
+
"serving_container": "qwen35-tt-bfp4-hf",
|
| 8 |
+
"serving_image": "qwen35-tt-bfp4:hf-downloaded",
|
| 9 |
+
"tt_metal_commit": "de59f8a658b1ceafd230c8266026b1a72bb198d7",
|
| 10 |
+
"vllm_commit": "03fa3af",
|
| 11 |
+
"loader": {
|
| 12 |
+
"input": "Original sharded BF16 safetensors + model.safetensors.index.json + composite config.json",
|
| 13 |
+
"keys": "model.language_model.* and lm_head.weight only; vision/MTP excluded",
|
| 14 |
+
"remap": "runtime/qwen36/tt/weight_mapping.py::remap_qwen36_state_dict",
|
| 15 |
+
"model_args": "Qwen36ModelArgs(mesh_device, max_batch_size=1, max_seq_len=128-rounded longest record)",
|
| 16 |
+
"construction": "Qwen36Model(mesh_device, args, remapped_state_dict, tensor_cache_path=args.weight_cache_path())",
|
| 17 |
+
"memory": "Evaluator memory-maps text weights through safetensors; avoids constructing a second full HF model. TT loaders still transpose/copy individual matrices and derive GDN mega/AB copies. BF16 weights plus derived device copies may approach P150 DRAM limits; requires device smoke, not assumed fit.",
|
| 18 |
+
"cache": "TT_CACHE_PATH=<workspace>/candidate-cache/<sha256-of-precision-runtime-config-source>; ModelArgs appends P150, then weight_cache_path appends dtype suffix. Never writes beneath original weights. Each runtime/precision map uses an independent content-addressed cache."
|
| 19 |
+
},
|
| 20 |
+
"architecture": {
|
| 21 |
+
"layers": 32,
|
| 22 |
+
"hidden_size": 4096,
|
| 23 |
+
"vocab_size": 248320,
|
| 24 |
+
"mlp_intermediate_size": 12288,
|
| 25 |
+
"full_attention_layers": [
|
| 26 |
+
3,
|
| 27 |
+
7,
|
| 28 |
+
11,
|
| 29 |
+
15,
|
| 30 |
+
19,
|
| 31 |
+
23,
|
| 32 |
+
27,
|
| 33 |
+
31
|
| 34 |
+
],
|
| 35 |
+
"gdn_layers": [
|
| 36 |
+
0,
|
| 37 |
+
1,
|
| 38 |
+
2,
|
| 39 |
+
4,
|
| 40 |
+
5,
|
| 41 |
+
6,
|
| 42 |
+
8,
|
| 43 |
+
9,
|
| 44 |
+
10,
|
| 45 |
+
12,
|
| 46 |
+
13,
|
| 47 |
+
14,
|
| 48 |
+
16,
|
| 49 |
+
17,
|
| 50 |
+
18,
|
| 51 |
+
20,
|
| 52 |
+
21,
|
| 53 |
+
22,
|
| 54 |
+
24,
|
| 55 |
+
25,
|
| 56 |
+
26,
|
| 57 |
+
28,
|
| 58 |
+
29,
|
| 59 |
+
30
|
| 60 |
+
]
|
| 61 |
+
},
|
| 62 |
+
"precision_inventory": {
|
| 63 |
+
"attention": {
|
| 64 |
+
"weights": [
|
| 65 |
+
"q_proj",
|
| 66 |
+
"k_proj",
|
| 67 |
+
"v_proj",
|
| 68 |
+
"o_proj"
|
| 69 |
+
],
|
| 70 |
+
"existing_default": "bfp8",
|
| 71 |
+
"current": "bfp4",
|
| 72 |
+
"generic_options": [
|
| 73 |
+
"bf16",
|
| 74 |
+
"bfp8",
|
| 75 |
+
"bfp4"
|
| 76 |
+
],
|
| 77 |
+
"granularity": "family per full-attention layer"
|
| 78 |
+
},
|
| 79 |
+
"gdn": {
|
| 80 |
+
"weights": [
|
| 81 |
+
"qkv_proj",
|
| 82 |
+
"in_proj_a",
|
| 83 |
+
"in_proj_b",
|
| 84 |
+
"in_proj_z",
|
| 85 |
+
"out_proj"
|
| 86 |
+
],
|
| 87 |
+
"existing_default": "bfp8",
|
| 88 |
+
"current": "bfp4",
|
| 89 |
+
"generic_options": [
|
| 90 |
+
"bf16",
|
| 91 |
+
"bfp8",
|
| 92 |
+
"bfp4"
|
| 93 |
+
],
|
| 94 |
+
"granularity": "family per GDN layer",
|
| 95 |
+
"derived": "AB and mega QKVABZ are concatenated from already TT-rounded component tensors then converted to the same TT dtype. This is the existing runtime loader behavior, not an approximation introduced by the evaluator."
|
| 96 |
+
},
|
| 97 |
+
"mlp_gate_up": {
|
| 98 |
+
"weights": [
|
| 99 |
+
"gate_proj",
|
| 100 |
+
"up_proj"
|
| 101 |
+
],
|
| 102 |
+
"existing_default": "hardcoded bfp4",
|
| 103 |
+
"current": "bfp4",
|
| 104 |
+
"generic_options": [
|
| 105 |
+
"bf16",
|
| 106 |
+
"bfp8",
|
| 107 |
+
"bfp4"
|
| 108 |
+
],
|
| 109 |
+
"granularity": "paired family per layer",
|
| 110 |
+
"change": "Isolated loader makes generic gate/up precision selectable; packed path remains BFP4-only"
|
| 111 |
+
},
|
| 112 |
+
"mlp_down": {
|
| 113 |
+
"weights": [
|
| 114 |
+
"down_proj"
|
| 115 |
+
],
|
| 116 |
+
"existing_default": "bfp8",
|
| 117 |
+
"current": "bfp4",
|
| 118 |
+
"generic_options": [
|
| 119 |
+
"bf16",
|
| 120 |
+
"bfp8",
|
| 121 |
+
"bfp4"
|
| 122 |
+
],
|
| 123 |
+
"granularity": "family per layer"
|
| 124 |
+
},
|
| 125 |
+
"lm_head": {
|
| 126 |
+
"weights": [
|
| 127 |
+
"output.weight"
|
| 128 |
+
],
|
| 129 |
+
"existing_default": "bfp8",
|
| 130 |
+
"current": "bfp4",
|
| 131 |
+
"generic_options": [
|
| 132 |
+
"bf16",
|
| 133 |
+
"bfp8",
|
| 134 |
+
"bfp4"
|
| 135 |
+
],
|
| 136 |
+
"granularity": "global"
|
| 137 |
+
},
|
| 138 |
+
"fixed_bf16": [
|
| 139 |
+
"token_embeddings",
|
| 140 |
+
"decoder RMSNorm weights",
|
| 141 |
+
"final RMSNorm",
|
| 142 |
+
"attention q_norm/k_norm",
|
| 143 |
+
"GDN conv taps/bias",
|
| 144 |
+
"GDN A_log/dt_bias/output norm",
|
| 145 |
+
"rotary tables",
|
| 146 |
+
"activations",
|
| 147 |
+
"KV caches",
|
| 148 |
+
"GDN recurrent and convolution state"
|
| 149 |
+
]
|
| 150 |
+
},
|
| 151 |
+
"profiles": {
|
| 152 |
+
"baseline-bf16": "BF16 for every linear family, including LM head and gate/up. Generic/unpacked TT operators and tiled BF16 recurrent state; conservative weight baseline, NOT full-precision HF arithmetic.",
|
| 153 |
+
"baseline-bfp8": "BFP8 for every linear family; generic runtime matched to baseline-bf16.",
|
| 154 |
+
"current-bfp4": "BFP4 for every linear family; generic runtime matched to other generic profiles. Not the packed serving numerical baseline.",
|
| 155 |
+
"serving-bfp4": "BFP4 weights and exact model-spec serving fusion/packed flags except MTP disabled and BF16 KV fixed. Uses the same _forward_decode core eagerly, no sampling/trace. This controls serving runtime numerics separately from generic BFP4; not a claimed end-to-end traced-serving parity result. Overrides rejected."
|
| 156 |
+
},
|
| 157 |
+
"override_schema": {
|
| 158 |
+
"families": {
|
| 159 |
+
"gdn": "bfp8",
|
| 160 |
+
"lm_head": "bf16"
|
| 161 |
+
},
|
| 162 |
+
"layers": {
|
| 163 |
+
"0": {
|
| 164 |
+
"gdn": "bf16",
|
| 165 |
+
"mlp_down": "bfp8"
|
| 166 |
+
},
|
| 167 |
+
"31": {
|
| 168 |
+
"attention": "bfp8",
|
| 169 |
+
"mlp_gate_up": "bfp8"
|
| 170 |
+
}
|
| 171 |
+
}
|
| 172 |
+
},
|
| 173 |
+
"override_precedence": "profile defaults < families < layers; layer indices must use canonical decimal strings; irrelevant layer-family combinations rejected",
|
| 174 |
+
"screening": "Use --input corpus.jsonl --split calibration --max-records N for calibration sweeps. Use --split heldout only after candidate selection. --max-records never truncates token IDs; all positions within every selected record are evaluated. Per-token wall time is printed and saved, including first-token compilation cost.",
|
| 175 |
+
"constraints": [
|
| 176 |
+
"No TT device run, compilation, build, test, formatter or linter was performed by this worker. Precision options are supported TT tensor dtypes on generic operators, but this exact model/profile still requires Main's device smoke and numerical validation.",
|
| 177 |
+
"Do not launch with device access while qwen35-tt-bfp4-hf or any other process owns the P150. This assignment does not authorize stopping/modifying it. --device-ownership-confirmed is an operator acknowledgement, not device arbitration.",
|
| 178 |
+
"Serving packed readers for LM head, MLP gate/up/down, GDN mega are BFP4-specific. Never feed BF16/BFP8 into those packed kernels.",
|
| 179 |
+
"BFP8 is quantized. BF16 weights do not remove BF16 state/activations, LoFi matmul configurations, softmax/norm approximation, or fused runtime effects. Score HF vs BF16 TT to establish runtime error; score generic BF16 TT vs matched generic quant candidates to isolate weight precision effects.",
|
| 180 |
+
"Do not compare serving-bfp4 vs generic mixed precision and attribute all KL difference to weights: native recurrence, frontend/output fusion, final residual norm, weight layout, and matmul dispatch differ.",
|
| 181 |
+
"Full vocabulary logits are read back before sampling; no top-k/top-p, no argmax substitute, no token skipping, no extra BOS/padding/EOS insertion.",
|
| 182 |
+
"Only original input token IDs are consumed; row p predicts token p+1. Final input token has no target and no emitted logit row. Calibration and heldout must remain distinct; never select precision using heldout.",
|
| 183 |
+
"Original HF files are read-only mmap inputs. Do not mutate original tensors in-place or write existing serving tensor caches.",
|
| 184 |
+
"No multimodal placeholders are expanded; this is a text-only evaluation contract."
|
| 185 |
+
],
|
| 186 |
+
"teacher_forcing": {
|
| 187 |
+
"path": "Allocate paged KV once, then Qwen36Model._forward_decode(tokens_tt, cos, sin, current_pos_tt, page_table) for every position 0..N-2; read output [*,248320] row zero as float32",
|
| 188 |
+
"first_position": "Position zero writes KV slot zero and consumes zero recurrent/convolution history; no prefill padding or history is needed by the inspected causal decode kernels. Runtime smoke required.",
|
| 189 |
+
"serving_reset": "_init_dn_zero_buffers once; _reset_dn_state_inplace per record clears recurrent tensors and shared backing convolution pool. Preserve native row-major BF16 L1 tensor addresses and views. Reset paged KV arrays too.",
|
| 190 |
+
"generic_reset": "Detach native state bindings, set use_inplace_state=False, initialize fresh tiled BF16 recurrent state in DRAM, clear all separate/fused/split conv histories. model.reset_state(batch_size=1) is not appropriate for this serving fork's native pool state.",
|
| 191 |
+
"artifacts": "metadata.json plus records.jsonl and per-record float32 NumPy .npy logits [N-1,248320]; include positions, token_ids, split, NLL, hashes, timings and exact precision/runtime metadata"
|
| 192 |
+
},
|
| 193 |
+
"serving_environment": {
|
| 194 |
+
"MESH_DEVICE": "P150",
|
| 195 |
+
"ARCH_NAME": "blackhole",
|
| 196 |
+
"TT_QWEN35_TEXT_VER": "qwen36_blackhole",
|
| 197 |
+
"QWEN36_MAX_TOKENS_ALL_USERS": "8192",
|
| 198 |
+
"QWEN36_BFP4_WEIGHTS": "1",
|
| 199 |
+
"QWEN36_BFP4_LM_HEAD": "1",
|
| 200 |
+
"QWEN36_LM_HEAD_PACKED": "1",
|
| 201 |
+
"QWEN36_LM_HEAD_REQUIRE_PACKED": "1",
|
| 202 |
+
"QWEN36_LM_HEAD_CB_SLOTS": "2",
|
| 203 |
+
"QWEN36_LM_HEAD_OUTSTANDING": "1",
|
| 204 |
+
"QWEN36_BFP4_MLP_DOWN": "1",
|
| 205 |
+
"QWEN36_MLP_DOWN_PACKED": "1",
|
| 206 |
+
"QWEN36_MLP_DOWN_REQUIRE_PACKED": "1",
|
| 207 |
+
"QWEN36_MLP_DOWN_CB_SLOTS": "2",
|
| 208 |
+
"QWEN36_MLP_DOWN_OUTSTANDING": "1",
|
| 209 |
+
"QWEN36_SINGLE_1D_DECODE": "1",
|
| 210 |
+
"QWEN36_FUSED_QKV": "1",
|
| 211 |
+
"QWEN36_SHARDED_FINAL_NORM": "1",
|
| 212 |
+
"QWEN36_FUSED_FINAL_RESIDUAL_NORM": "1",
|
| 213 |
+
"QWEN36_FUSED_GATE_UP": "1",
|
| 214 |
+
"QWEN36_FUSED_GATE_UP_PACKED": "1",
|
| 215 |
+
"QWEN36_GATEUP_REQUIRE_PACKED": "1",
|
| 216 |
+
"QWEN36_GATEUP_CB_SLOTS": "2",
|
| 217 |
+
"QWEN36_GATEUP_OUTSTANDING": "1",
|
| 218 |
+
"QWEN36_FUSED_GATE_UP_CORES": "110",
|
| 219 |
+
"QWEN_GDN_MEGA_CORES": "110",
|
| 220 |
+
"QWEN_GDN_MEGA_PACKED": "1",
|
| 221 |
+
"QWEN_GDN_MEGA_REQUIRE_PACKED": "1",
|
| 222 |
+
"QWEN_GDN_MEGA_CB_SLOTS": "2",
|
| 223 |
+
"QWEN_GDN_MEGA_OUTSTANDING": "1",
|
| 224 |
+
"QWEN_GDN_RECURRENT_FUSED": "1",
|
| 225 |
+
"QWEN_GDN_FRONTEND_FUSED": "1",
|
| 226 |
+
"QWEN_GDN_OUTPUT_FUSED": "1",
|
| 227 |
+
"QWEN36_MTP": "0",
|
| 228 |
+
"QWEN_SDPA_BF8": "0"
|
| 229 |
+
},
|
| 230 |
+
"safe_describe_command": "python3 /var/tmp/qwen9b-solid-quant/tt_eval.py --input /var/tmp/qwen9b-solid-quant/smoke.jsonl --weights /var/tmp/qwen9b-hf-publish/clean-cache/weights/Qwen3.5-9B --output /var/tmp/qwen9b-solid-quant/results/tt-bf16-smoke --profile baseline-bf16 --describe",
|
| 231 |
+
"device_launch_prerequisite": "Future exclusive P150 ownership explicitly established by Main/operator. Do NOT run this command now beside the live container.",
|
| 232 |
+
"device_launch_command": "docker run --rm --name qwen9b-tt-eval --network none --user 0:0 --device /dev/tenstorrent:/dev/tenstorrent --mount type=bind,src=/dev/hugepages,dst=/dev/hugepages --mount type=bind,src=/dev/hugepages-1G,dst=/dev/hugepages-1G --mount type=bind,src=/var/tmp/qwen9b-solid-quant,dst=/work --mount type=bind,src=/var/tmp/qwen9b-hf-publish/clean-cache/weights/Qwen3.5-9B,dst=/weights/Qwen3.5-9B,readonly --mount type=bind,src=/var/tmp/qwen9b-solid-quant/runtime/qwen36,dst=/home/container_app_user/tt-metal/models/demos/blackhole/qwen36,readonly -e PYTHONDONTWRITEBYTECODE=1 -e ARCH_NAME=blackhole -e MESH_DEVICE=P150 -e HOME=/work/eval-home -e XDG_CACHE_HOME=/work/eval-home/.cache -e TT_METAL_CACHE=/work/kernel-cache -e TT_METAL_LOGS_PATH=/work/eval-logs --entrypoint /home/container_app_user/tt-metal/python_env/bin/python qwen35-tt-bfp4:hf-downloaded /work/tt_eval.py --input /work/smoke.jsonl --weights /weights/Qwen3.5-9B --runtime /home/container_app_user/tt-metal/models/demos/blackhole/qwen36 --output /work/results/tt-bf16-smoke --profile baseline-bf16 --device-ownership-confirmed"
|
| 233 |
+
}
|
checkpoint/provenance/tt_eval.py
ADDED
|
@@ -0,0 +1,440 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
#!/usr/bin/env python3
|
| 2 |
+
"""Teacher-forced, full-vocabulary TT evaluation; never run beside a live TT server.
|
| 3 |
+
|
| 4 |
+
--describe validates inputs/precision and prints a launch description without
|
| 5 |
+
importing torch/ttnn or opening hardware. Device runs require explicit ownership.
|
| 6 |
+
Each N-token record produces N-1 rows: row p predicts token_ids[p+1].
|
| 7 |
+
"""
|
| 8 |
+
import argparse
|
| 9 |
+
import datetime
|
| 10 |
+
import gc
|
| 11 |
+
import hashlib
|
| 12 |
+
import json
|
| 13 |
+
import os
|
| 14 |
+
from pathlib import Path
|
| 15 |
+
import sys
|
| 16 |
+
import time
|
| 17 |
+
|
| 18 |
+
ROOT = Path(__file__).resolve().parent
|
| 19 |
+
REVISION = "c202236235762e1c871ad0ccb60c8ee5ba337b9a"
|
| 20 |
+
FAMILIES = {"attention", "gdn", "gdn_output", "mlp_gate_up", "mlp_down", "lm_head"}
|
| 21 |
+
DTYPES = {"bf16", "bfp8", "bfp4"}
|
| 22 |
+
PROFILES = {"baseline-bf16": "bf16", "baseline-bfp8": "bfp8", "current-bfp4": "bfp4", "serving-bfp4": "bfp4", "mixed-packed": "bfp4"}
|
| 23 |
+
|
| 24 |
+
|
| 25 |
+
def digest(path):
|
| 26 |
+
h = hashlib.sha256()
|
| 27 |
+
with Path(path).open("rb") as stream:
|
| 28 |
+
for block in iter(lambda: stream.read(1024 * 1024), b""):
|
| 29 |
+
h.update(block)
|
| 30 |
+
return h.hexdigest()
|
| 31 |
+
|
| 32 |
+
|
| 33 |
+
def canonical(value):
|
| 34 |
+
return json.dumps(value, sort_keys=True, separators=(",", ":"))
|
| 35 |
+
|
| 36 |
+
|
| 37 |
+
def save_json(path, value):
|
| 38 |
+
temporary = path.with_suffix(path.suffix + ".partial")
|
| 39 |
+
temporary.write_text(json.dumps(value, indent=2, sort_keys=True) + "\n")
|
| 40 |
+
temporary.replace(path)
|
| 41 |
+
|
| 42 |
+
|
| 43 |
+
def precision_map(profile, overrides, layer_types):
|
| 44 |
+
default = PROFILES[profile]
|
| 45 |
+
result = {"lm_head": default, "layers": {}}
|
| 46 |
+
for index, kind in enumerate(layer_types):
|
| 47 |
+
attention = "attention" if kind == "full_attention" else "gdn"
|
| 48 |
+
result["layers"][str(index)] = {attention: default, "mlp_gate_up": default, "mlp_down": default}
|
| 49 |
+
if overrides is None:
|
| 50 |
+
return result
|
| 51 |
+
if profile == "serving-bfp4":
|
| 52 |
+
raise ValueError("serving-bfp4 fixes the packed runtime and does not permit precision overrides")
|
| 53 |
+
if not isinstance(overrides, dict) or set(overrides) - {"families", "layers"}:
|
| 54 |
+
raise ValueError("Overrides must contain only families and/or layers objects")
|
| 55 |
+
families = overrides.get("families", {})
|
| 56 |
+
if not isinstance(families, dict):
|
| 57 |
+
raise ValueError("families must be an object")
|
| 58 |
+
for family, dtype in families.items():
|
| 59 |
+
if family not in FAMILIES or dtype not in DTYPES:
|
| 60 |
+
raise ValueError(f"Unsupported family/dtype: {family}={dtype}")
|
| 61 |
+
if family == "lm_head":
|
| 62 |
+
result[family] = dtype
|
| 63 |
+
else:
|
| 64 |
+
for values in result["layers"].values():
|
| 65 |
+
if family in values or (family == "gdn_output" and "gdn" in values):
|
| 66 |
+
values[family] = dtype
|
| 67 |
+
layers = overrides.get("layers", {})
|
| 68 |
+
if not isinstance(layers, dict):
|
| 69 |
+
raise ValueError("layers must be an object keyed by canonical layer numbers")
|
| 70 |
+
for index, values in layers.items():
|
| 71 |
+
if index not in result["layers"] or not isinstance(values, dict):
|
| 72 |
+
raise ValueError(f"Unknown layer or non-object precision override: {index}")
|
| 73 |
+
for family, dtype in values.items():
|
| 74 |
+
if (family not in result["layers"][index] and not (family == "gdn_output" and "gdn" in result["layers"][index])) or dtype not in DTYPES:
|
| 75 |
+
raise ValueError(f"Unsupported layer family/dtype: {index}.{family}={dtype}")
|
| 76 |
+
result["layers"][index][family] = dtype
|
| 77 |
+
return result
|
| 78 |
+
|
| 79 |
+
|
| 80 |
+
def load_records(path, vocab_size, context_limit):
|
| 81 |
+
records, seen, split_tokens = [], set(), {}
|
| 82 |
+
with path.open() as stream:
|
| 83 |
+
for number, line in enumerate(stream, 1):
|
| 84 |
+
if not line.strip():
|
| 85 |
+
continue
|
| 86 |
+
row = json.loads(line)
|
| 87 |
+
if not isinstance(row, dict) or not {"id", "split", "token_ids"} <= row.keys():
|
| 88 |
+
raise ValueError(f"Record {number}: expected id, split, token_ids")
|
| 89 |
+
key, split, tokens = row["id"], row["split"], row["token_ids"]
|
| 90 |
+
if isinstance(key, bool) or not isinstance(key, (str, int)) or canonical(key) in seen:
|
| 91 |
+
raise ValueError(f"Record {number}: id must be a unique string/integer")
|
| 92 |
+
if split not in {"calibration", "validation", "heldout"}:
|
| 93 |
+
raise ValueError(f"Record {number}: split must be calibration, validation or heldout")
|
| 94 |
+
if not isinstance(tokens, list) or not 2 <= len(tokens) <= context_limit:
|
| 95 |
+
raise ValueError(f"Record {number}: token_ids length must be in [2,{context_limit}]")
|
| 96 |
+
if any(type(token) is not int or not 0 <= token < vocab_size for token in tokens):
|
| 97 |
+
raise ValueError(f"Record {number}: token ID outside vocabulary")
|
| 98 |
+
token_hash = hashlib.sha256(canonical(tokens).encode()).hexdigest()
|
| 99 |
+
if token_hash in split_tokens and split_tokens[token_hash] != split:
|
| 100 |
+
raise ValueError("Identical token sequence appears in different splits")
|
| 101 |
+
split_tokens[token_hash] = split
|
| 102 |
+
seen.add(canonical(key))
|
| 103 |
+
records.append({"id": key, "split": split, "token_ids": tokens})
|
| 104 |
+
if not records:
|
| 105 |
+
raise ValueError("Input dataset is empty")
|
| 106 |
+
return records
|
| 107 |
+
|
| 108 |
+
|
| 109 |
+
def runtime_environment(profile, inventory, mtp):
|
| 110 |
+
if profile in ("serving-bfp4", "mixed-packed"):
|
| 111 |
+
values = dict(inventory["serving_environment"])
|
| 112 |
+
else:
|
| 113 |
+
values = {
|
| 114 |
+
"MESH_DEVICE": "P150", "ARCH_NAME": "blackhole",
|
| 115 |
+
"QWEN36_SINGLE_1D_DECODE": "1",
|
| 116 |
+
"QWEN_GDN_RECURRENT_FUSED": "1",
|
| 117 |
+
"QWEN_GDN_FRONTEND_FUSED": "1",
|
| 118 |
+
}
|
| 119 |
+
values.update({"QWEN36_MTP": "1" if mtp else "0", "QWEN_SDPA_BF8": "0"})
|
| 120 |
+
return values
|
| 121 |
+
|
| 122 |
+
|
| 123 |
+
def load_text_weights(weights, torch, mtp):
|
| 124 |
+
"""Memory-map original text weights and optional lossless MTP; no HF allocation."""
|
| 125 |
+
from safetensors import safe_open
|
| 126 |
+
from models.demos.blackhole.qwen36.tt.weight_mapping import remap_qwen36_state_dict
|
| 127 |
+
|
| 128 |
+
if (weights / "native_manifest.json").is_file():
|
| 129 |
+
from native_checkpoint import load_native_checkpoint
|
| 130 |
+
return load_native_checkpoint(weights, allow_unverified=True)
|
| 131 |
+
index = json.loads((weights / "model.safetensors.index.json").read_text())["weight_map"]
|
| 132 |
+
by_file = {}
|
| 133 |
+
for name, filename in index.items():
|
| 134 |
+
if name.startswith("model.language_model.") or name == "lm_head.weight":
|
| 135 |
+
by_file.setdefault(filename, []).append(name)
|
| 136 |
+
if not by_file:
|
| 137 |
+
raise ValueError("Expected original Qwen3.5 multimodal checkpoint text-weight names")
|
| 138 |
+
raw = {}
|
| 139 |
+
for filename, names in sorted(by_file.items()):
|
| 140 |
+
shard = (weights / filename).resolve()
|
| 141 |
+
if not shard.is_relative_to(weights):
|
| 142 |
+
raise ValueError("Checkpoint index shard escapes read-only weights directory")
|
| 143 |
+
with safe_open(str(shard), framework="pt", device="cpu") as source:
|
| 144 |
+
for name in names:
|
| 145 |
+
tensor = source.get_tensor(name)
|
| 146 |
+
expected = torch.float32 if name.endswith((".linear_attn.A_log", ".linear_attn.norm.weight")) else torch.bfloat16
|
| 147 |
+
if tensor.dtype != expected:
|
| 148 |
+
raise ValueError(f"Expected original {expected} weight, got {name}: {tensor.dtype}")
|
| 149 |
+
raw[name] = tensor
|
| 150 |
+
remapped = remap_qwen36_state_dict(raw)
|
| 151 |
+
if not {"tok_embeddings.weight", "output.weight", "norm.weight"} <= remapped.keys():
|
| 152 |
+
raise ValueError("Text checkpoint is missing embedding/head/norm weights")
|
| 153 |
+
if mtp:
|
| 154 |
+
from models.demos.blackhole.qwen36.tt.weight_mapping import load_qwen36_mtp_state_dict
|
| 155 |
+
|
| 156 |
+
remapped.update(load_qwen36_mtp_state_dict(weights))
|
| 157 |
+
return remapped
|
| 158 |
+
|
| 159 |
+
|
| 160 |
+
def reset_sequence(model, kv_caches, kv_zero, native, ttnn):
|
| 161 |
+
"""Reset all causal history, preserving native GDN pool views when applicable."""
|
| 162 |
+
model.rope.rope_delta = 0
|
| 163 |
+
model._req_image_grid_thw = None
|
| 164 |
+
model._req_video_grid_thw = None
|
| 165 |
+
if model._last_hidden is not None:
|
| 166 |
+
ttnn.deallocate(model._last_hidden)
|
| 167 |
+
model._last_hidden = None
|
| 168 |
+
if model.mtp is not None:
|
| 169 |
+
model.mtp.reset_cache()
|
| 170 |
+
for caches in kv_caches:
|
| 171 |
+
for cache in caches:
|
| 172 |
+
ttnn.copy(kv_zero, cache)
|
| 173 |
+
if native:
|
| 174 |
+
# This zeros the backing conv pool ONCE, not per-layer alias views.
|
| 175 |
+
model._reset_dn_state_inplace()
|
| 176 |
+
return
|
| 177 |
+
for layer in model.layers:
|
| 178 |
+
if layer.is_full_attention:
|
| 179 |
+
layer.attention.reset_cache()
|
| 180 |
+
continue
|
| 181 |
+
dn = layer.attention
|
| 182 |
+
# Generic TT recurrence consumes tiled state, not the native row-major
|
| 183 |
+
# L1 pools allocated by this serving fork. Do not use model.reset_state:
|
| 184 |
+
# it leaves use_inplace_state=True and drops external convolution views.
|
| 185 |
+
dn.use_inplace_state = False
|
| 186 |
+
dn._chunk_inplace_state = False
|
| 187 |
+
dn.recurrent_state = ttnn.zeros(
|
| 188 |
+
[1, dn.num_v_heads, dn.head_k_dim, dn.head_v_dim],
|
| 189 |
+
dtype=ttnn.bfloat16, layout=ttnn.TILE_LAYOUT,
|
| 190 |
+
device=model.device, memory_config=ttnn.DRAM_MEMORY_CONFIG,
|
| 191 |
+
)
|
| 192 |
+
dn.conv_state_q = dn.conv_state_k = dn.conv_state_v = None
|
| 193 |
+
dn.fused_conv_state = None
|
| 194 |
+
dn.split_conv_state = None
|
| 195 |
+
gc.collect()
|
| 196 |
+
|
| 197 |
+
|
| 198 |
+
def evaluate(args, records, metadata, cache, environment):
|
| 199 |
+
# Scrub inherited serving flags BEFORE importing model/experimental modules.
|
| 200 |
+
for name in list(os.environ):
|
| 201 |
+
if name.startswith(("QWEN", "TT_QWEN")):
|
| 202 |
+
del os.environ[name]
|
| 203 |
+
os.environ.update(environment)
|
| 204 |
+
os.environ.update({
|
| 205 |
+
"HF_MODEL": str(args.weights), "MODEL_WEIGHTS_DIR": str(args.weights),
|
| 206 |
+
"TT_CACHE_PATH": str(cache), "HF_HUB_OFFLINE": "1", "TRANSFORMERS_OFFLINE": "1",
|
| 207 |
+
"QWEN36_EVAL_PRECISION": canonical(metadata["precision"]),
|
| 208 |
+
"TT_METAL_CACHE": str(cache / "kernel-cache"),
|
| 209 |
+
})
|
| 210 |
+
import torch
|
| 211 |
+
import numpy as np
|
| 212 |
+
import ttnn
|
| 213 |
+
from models.demos.blackhole.qwen36.tt.model import Qwen36Model
|
| 214 |
+
from models.demos.blackhole.qwen36.tt.model_config import Qwen36ModelArgs
|
| 215 |
+
|
| 216 |
+
imported_source = Path(sys.modules[Qwen36Model.__module__].__file__).resolve()
|
| 217 |
+
expected_source = args.runtime.resolve() / "tt" / "model.py"
|
| 218 |
+
if imported_source != expected_source:
|
| 219 |
+
raise RuntimeError(f"Refusing non-isolated runtime: imported {imported_source}; expected {expected_source}")
|
| 220 |
+
torch.set_num_threads(args.cpu_threads)
|
| 221 |
+
mesh = ttnn.open_mesh_device(
|
| 222 |
+
mesh_shape=ttnn.MeshShape(1, 1), l1_small_size=24576,
|
| 223 |
+
num_command_queues=2, trace_region_size=0,
|
| 224 |
+
)
|
| 225 |
+
try:
|
| 226 |
+
if mesh.get_num_devices() != 1:
|
| 227 |
+
raise RuntimeError("Evaluator supports one P150 only")
|
| 228 |
+
model_args = Qwen36ModelArgs(mesh_device=mesh, max_batch_size=1, max_seq_len=metadata["max_seq_len"])
|
| 229 |
+
resolved_cache = model_args.weight_cache_path().resolve()
|
| 230 |
+
if not resolved_cache.is_relative_to(cache):
|
| 231 |
+
raise RuntimeError(f"Unsafe tensor cache path: {resolved_cache}")
|
| 232 |
+
resolved_cache.mkdir(parents=True, exist_ok=True)
|
| 233 |
+
state_dict = load_text_weights(args.weights, torch, args.mtp)
|
| 234 |
+
model = Qwen36Model(mesh, model_args, state_dict, tensor_cache_path=resolved_cache)
|
| 235 |
+
del state_dict
|
| 236 |
+
gc.collect()
|
| 237 |
+
if (model.mtp is not None) != args.mtp or model.vocab_size != metadata["vocab_size"]:
|
| 238 |
+
raise RuntimeError("MTP/vocabulary contract violation")
|
| 239 |
+
model._ondev_argmax = False
|
| 240 |
+
native = environment.get("QWEN_GDN_RECURRENT_FUSED") == "1"
|
| 241 |
+
blocks = metadata["max_seq_len"] // 64
|
| 242 |
+
kv_shape = [blocks, model_args.n_kv_heads, 64, model_args.head_dim]
|
| 243 |
+
kv_caches = model.allocate_kv_caches(kv_shape, ttnn.bfloat16, batch_size=1)
|
| 244 |
+
kv_zero = ttnn.zeros(kv_shape, dtype=ttnn.bfloat16, layout=ttnn.TILE_LAYOUT, device=mesh)
|
| 245 |
+
page_table = ttnn.from_torch(torch.arange(blocks, dtype=torch.int32).reshape(1, -1),
|
| 246 |
+
dtype=ttnn.int32, layout=ttnn.ROW_MAJOR_LAYOUT, device=mesh)
|
| 247 |
+
if native:
|
| 248 |
+
model._init_dn_zero_buffers()
|
| 249 |
+
metadata.update({"torch_version": torch.__version__, "ttnn_version": getattr(ttnn, "__version__", None),
|
| 250 |
+
"tensor_cache_path": str(resolved_cache), "imported_model_source": str(imported_source),
|
| 251 |
+
"device_grid": str(mesh.compute_with_storage_grid_size())})
|
| 252 |
+
save_json(args.output / "metadata.json", metadata)
|
| 253 |
+
generations = []
|
| 254 |
+
tokenizer = None
|
| 255 |
+
if args.generate_tokens:
|
| 256 |
+
from transformers import AutoTokenizer
|
| 257 |
+
tokenizer = AutoTokenizer.from_pretrained(args.weights, local_files_only=True)
|
| 258 |
+
with torch.inference_mode(), (args.output / "records.jsonl").open("x") as manifest:
|
| 259 |
+
for record_index, record in enumerate(records):
|
| 260 |
+
started = time.monotonic()
|
| 261 |
+
reset_sequence(model, kv_caches, kv_zero, native, ttnn)
|
| 262 |
+
prompt_token_ids = list(record["token_ids"])
|
| 263 |
+
token_ids = list(prompt_token_ids)
|
| 264 |
+
count = len(token_ids) - 1 + args.generate_tokens
|
| 265 |
+
filename = f"logits-{record_index:06d}.npy"
|
| 266 |
+
target_path = args.output / filename
|
| 267 |
+
temporary_path = args.output / (filename + ".partial")
|
| 268 |
+
logits_array = np.lib.format.open_memmap(temporary_path, mode="w+", dtype=np.float32,
|
| 269 |
+
shape=(count, model.vocab_size))
|
| 270 |
+
nll_sum = 0.0
|
| 271 |
+
token_seconds = []
|
| 272 |
+
for position in range(count):
|
| 273 |
+
token = token_ids[position]
|
| 274 |
+
token_started = time.monotonic()
|
| 275 |
+
tokens_tt = ttnn.from_torch(torch.tensor([[token]], dtype=torch.int32),
|
| 276 |
+
dtype=ttnn.uint32, layout=ttnn.ROW_MAJOR_LAYOUT, device=mesh)
|
| 277 |
+
position_tt = ttnn.from_torch(torch.tensor([position], dtype=torch.int32),
|
| 278 |
+
dtype=ttnn.int32, layout=ttnn.ROW_MAJOR_LAYOUT, device=mesh)
|
| 279 |
+
cos, sin = model.rope.get_rot_mats(torch.tensor([[position]], dtype=torch.long))
|
| 280 |
+
output = model._forward_decode(tokens_tt, cos, sin, position_tt, page_table)
|
| 281 |
+
host = ttnn.to_torch(output).float()
|
| 282 |
+
if host.shape[-1] != model.vocab_size or host.numel() < model.vocab_size:
|
| 283 |
+
raise RuntimeError(f"Invalid full-vocabulary logits shape: {tuple(host.shape)}")
|
| 284 |
+
row = host.reshape(-1, model.vocab_size)[0]
|
| 285 |
+
if not torch.isfinite(row).all().item():
|
| 286 |
+
raise RuntimeError(f"Non-finite logits at record {record['id']}, position {position}")
|
| 287 |
+
logits_array[position] = row.numpy()
|
| 288 |
+
stop_generation = False
|
| 289 |
+
if args.generate_tokens and position >= len(prompt_token_ids) - 1:
|
| 290 |
+
sampled = int(row.argmax())
|
| 291 |
+
token_ids.append(sampled)
|
| 292 |
+
stop_generation = sampled == tokenizer.eos_token_id
|
| 293 |
+
nll_sum += float(torch.logsumexp(row, dim=-1) - row[token_ids[position + 1]])
|
| 294 |
+
# RoPE slices can alias persistent tables: let references
|
| 295 |
+
# release normally instead of force-deallocating their owner.
|
| 296 |
+
for tensor in (output, tokens_tt, position_tt):
|
| 297 |
+
ttnn.deallocate(tensor)
|
| 298 |
+
token_seconds.append(time.monotonic() - token_started)
|
| 299 |
+
print(canonical({"record": record["id"], "position": position,
|
| 300 |
+
"token_elapsed_seconds": token_seconds[-1]}), flush=True)
|
| 301 |
+
if stop_generation:
|
| 302 |
+
count = position + 1
|
| 303 |
+
break
|
| 304 |
+
logits_array.flush()
|
| 305 |
+
if count < logits_array.shape[0]:
|
| 306 |
+
trimmed = temporary_path.with_suffix(".trimmed")
|
| 307 |
+
with trimmed.open("wb") as stream:
|
| 308 |
+
np.save(stream, logits_array[:count])
|
| 309 |
+
trimmed.replace(temporary_path)
|
| 310 |
+
del logits_array
|
| 311 |
+
temporary_path.replace(target_path)
|
| 312 |
+
item = {**record, "token_ids": token_ids, "positions": list(range(count)), "logits_file": filename,
|
| 313 |
+
"logits_shape": [count, model.vocab_size], "logits_dtype": "float32",
|
| 314 |
+
"nll_sum": nll_sum, "nll_mean": nll_sum / count, "nll_tokens": count,
|
| 315 |
+
"token_elapsed_seconds": token_seconds,
|
| 316 |
+
"decode_seconds_per_token": sum(token_seconds) / count,
|
| 317 |
+
"elapsed_seconds": time.monotonic() - started, "logits_sha256": digest(target_path)}
|
| 318 |
+
manifest.write(canonical(item) + "\n")
|
| 319 |
+
manifest.flush()
|
| 320 |
+
if args.generate_tokens:
|
| 321 |
+
generated = token_ids[len(prompt_token_ids):]
|
| 322 |
+
generations.append({"id": record["id"], "prompt_token_ids": prompt_token_ids,
|
| 323 |
+
"generated_token_ids": generated,
|
| 324 |
+
"text": tokenizer.decode(generated, skip_special_tokens=False),
|
| 325 |
+
"seconds": time.monotonic() - started})
|
| 326 |
+
print(canonical({"record": record["id"], "split": record["split"], "nll_mean": item["nll_mean"]}), flush=True)
|
| 327 |
+
if args.generate_tokens:
|
| 328 |
+
(args.output / "generations.jsonl").write_text("".join(canonical(row) + "\n" for row in generations))
|
| 329 |
+
metadata["status"] = "complete"
|
| 330 |
+
metadata["completed_records"] = len(records)
|
| 331 |
+
finally:
|
| 332 |
+
ttnn.close_mesh_device(mesh)
|
| 333 |
+
|
| 334 |
+
|
| 335 |
+
def main():
|
| 336 |
+
parser = argparse.ArgumentParser(description=__doc__)
|
| 337 |
+
parser.add_argument("--input", type=Path, required=True)
|
| 338 |
+
parser.add_argument("--weights", type=Path, required=True)
|
| 339 |
+
parser.add_argument("--output", type=Path, required=True)
|
| 340 |
+
parser.add_argument("--cache-root", type=Path, default=ROOT / "candidate-cache")
|
| 341 |
+
parser.add_argument("--runtime", type=Path, default=ROOT / "runtime" / "qwen36")
|
| 342 |
+
parser.add_argument("--profile", choices=sorted(PROFILES), default="baseline-bf16")
|
| 343 |
+
parser.add_argument("--overrides", type=Path)
|
| 344 |
+
parser.add_argument("--context-limit", type=int, default=8192)
|
| 345 |
+
parser.add_argument("--cpu-threads", type=int, default=8)
|
| 346 |
+
parser.add_argument("--mtp", action="store_true",
|
| 347 |
+
help="Load the MTP head while evaluating full target logits; does not run speculative generation")
|
| 348 |
+
parser.add_argument("--generate-tokens", type=int, default=0,
|
| 349 |
+
help="Free greedy generation diagnostics instead of fixed-input teacher forcing")
|
| 350 |
+
parser.add_argument("--split", choices=("calibration", "validation", "heldout"),
|
| 351 |
+
help="Select one split; screen candidates on calibration only")
|
| 352 |
+
parser.add_argument("--max-records", type=int,
|
| 353 |
+
help="Evaluate only the first N records after split filtering; never truncates tokens")
|
| 354 |
+
parser.add_argument("--describe", action="store_true")
|
| 355 |
+
parser.add_argument("--device-ownership-confirmed", action="store_true",
|
| 356 |
+
help="Required for device execution; operator must first ensure no other process owns the P150")
|
| 357 |
+
args = parser.parse_args()
|
| 358 |
+
for name in ("weights", "input", "output", "cache_root", "runtime"):
|
| 359 |
+
setattr(args, name, getattr(args, name).resolve())
|
| 360 |
+
if not 2 <= args.context_limit <= 8192 or args.cpu_threads < 1:
|
| 361 |
+
raise ValueError("context-limit must be 2..8192 and cpu-threads positive")
|
| 362 |
+
if not 0 <= args.generate_tokens <= 256:
|
| 363 |
+
raise ValueError("generate-tokens must be in [0,256]")
|
| 364 |
+
# Restrict every persistent artifact/cache write to this isolated workspace.
|
| 365 |
+
for path in (args.output, args.cache_root):
|
| 366 |
+
if not path.is_relative_to(ROOT) or path == ROOT or path.is_relative_to(args.weights):
|
| 367 |
+
raise ValueError(f"Write path must be isolated under {ROOT}, never under weights: {path}")
|
| 368 |
+
config_path = args.weights / "config.json"
|
| 369 |
+
config = json.loads(config_path.read_text())["text_config"]
|
| 370 |
+
if config["num_hidden_layers"] != 32 or config["hidden_size"] != 4096 or config["vocab_size"] != 248320:
|
| 371 |
+
raise ValueError("This evaluator is scoped to the original Qwen3.5-9B architecture")
|
| 372 |
+
records = load_records(args.input, config["vocab_size"], args.context_limit)
|
| 373 |
+
if args.split:
|
| 374 |
+
records = [record for record in records if record["split"] == args.split]
|
| 375 |
+
if args.max_records is not None:
|
| 376 |
+
if args.max_records < 1:
|
| 377 |
+
raise ValueError("max-records must be positive")
|
| 378 |
+
records = records[:args.max_records]
|
| 379 |
+
if not records:
|
| 380 |
+
raise ValueError("No records remain after split/record selection")
|
| 381 |
+
if max(len(record["token_ids"]) for record in records) + args.generate_tokens > args.context_limit:
|
| 382 |
+
raise ValueError("Prompt plus generation exceeds the declared context limit")
|
| 383 |
+
inventory = json.loads((ROOT / "precision-plan.json").read_text())
|
| 384 |
+
overrides = json.loads(args.overrides.read_text()) if args.overrides else None
|
| 385 |
+
precision = precision_map(args.profile, overrides, config["layer_types"])
|
| 386 |
+
environment = runtime_environment(args.profile, inventory, args.mtp)
|
| 387 |
+
source_hashes = {str(path.relative_to(args.runtime)): digest(path) for path in sorted(args.runtime.rglob("*"))
|
| 388 |
+
if path.is_file() and path.suffix in {".py", ".cpp", ".hpp", ".h"}}
|
| 389 |
+
identity = {"precision": precision, "environment": environment, "sources": source_hashes,
|
| 390 |
+
"config_sha256": digest(config_path), "declared_upstream_revision": REVISION}
|
| 391 |
+
native_manifest_path = args.weights / "native_manifest.json"
|
| 392 |
+
if native_manifest_path.is_file():
|
| 393 |
+
native_manifest = json.loads(native_manifest_path.read_text())
|
| 394 |
+
if native_manifest["precision"] != precision:
|
| 395 |
+
raise ValueError("Requested precision differs from the native checkpoint")
|
| 396 |
+
identity["native_manifest_sha256"] = digest(native_manifest_path)
|
| 397 |
+
else:
|
| 398 |
+
identity["index_sha256"] = digest(args.weights / "model.safetensors.index.json")
|
| 399 |
+
candidate_hash = hashlib.sha256(canonical(identity).encode()).hexdigest()
|
| 400 |
+
cache = args.cache_root / candidate_hash
|
| 401 |
+
metadata = {"schema_version": 1, "backend": "ttnn-native", "status": "not_started",
|
| 402 |
+
"profile": args.profile, "precision": precision, "candidate_sha256": candidate_hash,
|
| 403 |
+
"input_sha256": digest(args.input), "vocab_size": config["vocab_size"],
|
| 404 |
+
"selection": {"split": args.split, "max_records": args.max_records,
|
| 405 |
+
"record_ids": [record["id"] for record in records]},
|
| 406 |
+
"max_seq_len": max(128, ((max(len(r["token_ids"]) for r in records) + args.generate_tokens + 127) // 128) * 128),
|
| 407 |
+
"source_identity": identity, "cache_path": str(cache), "runtime_environment": environment,
|
| 408 |
+
"alignment": "row p consumes token_ids[:p+1], predicts token_ids[p+1]; positions 0..N-2",
|
| 409 |
+
"full_vocabulary": True, "mtp": args.mtp, "vision": False, "trace": False,
|
| 410 |
+
"speculative_generation": False,
|
| 411 |
+
"generation_max_new_tokens": args.generate_tokens,
|
| 412 |
+
"likelihood_scope": "prompt targets plus self-selected greedy targets; not heldout NLL" if args.generate_tokens else "fixed teacher-forced input targets",
|
| 413 |
+
"state_dtype": "bf16", "activation_dtype": "bf16", "kv_dtype": "bf16",
|
| 414 |
+
"weight_loader": "native quantized checkpoint, evaluation-only unverified load" if native_manifest_path.is_file() else "read-only original text safetensors, preserving FP32 exceptions and optional lossless MTP",
|
| 415 |
+
"evaluator_sha256": digest(Path(__file__)),
|
| 416 |
+
"runtime_comparison_caveat": "BF16 weights retain BF16 arithmetic and native GDN recurrence/frontend. Generic controls share nonpacked matmul settings; packed/output/norm fusions are a separate runtime comparison.",
|
| 417 |
+
"created_at": datetime.datetime.now(datetime.timezone.utc).isoformat()}
|
| 418 |
+
if args.describe:
|
| 419 |
+
print(json.dumps(metadata, indent=2, sort_keys=True))
|
| 420 |
+
return
|
| 421 |
+
if not args.device_ownership_confirmed:
|
| 422 |
+
raise RuntimeError("Refusing device access: use --describe now. Execution requires exclusive P150 ownership, not merely a second container.")
|
| 423 |
+
if args.output.exists() and any(args.output.iterdir()):
|
| 424 |
+
raise ValueError("Output directory must be new/empty; refusing to overwrite prior evaluation")
|
| 425 |
+
args.output.mkdir(parents=True, exist_ok=True)
|
| 426 |
+
cache.mkdir(parents=True, exist_ok=True)
|
| 427 |
+
metadata["status"] = "running"
|
| 428 |
+
save_json(args.output / "metadata.json", metadata)
|
| 429 |
+
try:
|
| 430 |
+
evaluate(args, records, metadata, cache, environment)
|
| 431 |
+
except BaseException as error:
|
| 432 |
+
metadata["status"] = "failed"
|
| 433 |
+
metadata["error"] = f"{type(error).__name__}: {error}"
|
| 434 |
+
raise
|
| 435 |
+
finally:
|
| 436 |
+
save_json(args.output / "metadata.json", metadata)
|
| 437 |
+
|
| 438 |
+
|
| 439 |
+
if __name__ == "__main__":
|
| 440 |
+
main()
|
checkpoint/provenance/weight_mapping.py
ADDED
|
@@ -0,0 +1,220 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# SPDX-FileCopyrightText: © 2026 Tenstorrent USA, Inc.
|
| 2 |
+
# SPDX-License-Identifier: Apache-2.0
|
| 3 |
+
|
| 4 |
+
"""Remap HuggingFace Qwen3.5-9B state dict to internal format.
|
| 5 |
+
|
| 6 |
+
Handles:
|
| 7 |
+
- Stripping 'model.language_model.' prefix
|
| 8 |
+
- Filtering out vision encoder and MTP weights
|
| 9 |
+
- Renaming combined in_proj_qkv → qkv_proj (DeltaNet layers; the op uses the fused weight)
|
| 10 |
+
- Splitting combined conv1d.weight into separate Q, K, V conv weights (DeltaNet layers)
|
| 11 |
+
- Renaming lm_head.weight → output.weight
|
| 12 |
+
- Renaming embed_tokens → tok_embeddings
|
| 13 |
+
"""
|
| 14 |
+
import json
|
| 15 |
+
from pathlib import Path
|
| 16 |
+
from typing import Dict
|
| 17 |
+
|
| 18 |
+
import torch
|
| 19 |
+
|
| 20 |
+
# Layer indices that use full (softmax) attention
|
| 21 |
+
FULL_ATTENTION_LAYERS = {3, 7, 11, 15, 19, 23, 27, 31}
|
| 22 |
+
|
| 23 |
+
# DeltaNet QKV split dimensions (used to split the combined conv1d.weight into
|
| 24 |
+
# per-stream Q/K/V conv weights — the QKV projection itself stays combined).
|
| 25 |
+
# Q: num_key_heads(16) × key_head_dim(128) = 2048
|
| 26 |
+
# K: num_key_heads(16) × key_head_dim(128) = 2048
|
| 27 |
+
# (V = num_value_heads(32) × value_head_dim(128) = 4096 is the remaining slice)
|
| 28 |
+
LINEAR_Q_DIM = 2048
|
| 29 |
+
LINEAR_K_DIM = 2048
|
| 30 |
+
|
| 31 |
+
|
| 32 |
+
def remap_qwen36_state_dict(state_dict: Dict[str, torch.Tensor]) -> Dict[str, torch.Tensor]:
|
| 33 |
+
"""Remap HF Qwen3.5-9B state dict to internal format.
|
| 34 |
+
|
| 35 |
+
Args:
|
| 36 |
+
state_dict: Raw HuggingFace state dict loaded from safetensors.
|
| 37 |
+
|
| 38 |
+
Returns:
|
| 39 |
+
Remapped state dict with internal naming convention.
|
| 40 |
+
"""
|
| 41 |
+
remapped = {}
|
| 42 |
+
|
| 43 |
+
for key, tensor in state_dict.items():
|
| 44 |
+
# Filter out vision encoder weights (check original key — no prefix stripping yet)
|
| 45 |
+
if "visual" in key or key.startswith("model.visual"):
|
| 46 |
+
continue
|
| 47 |
+
# Filter out MTP (multi-token prediction) weights (original key)
|
| 48 |
+
if key.startswith("mtp"):
|
| 49 |
+
continue
|
| 50 |
+
|
| 51 |
+
# Strip the language-model prefix. Two checkpoint sources produce different
|
| 52 |
+
# prefixes for the same internal weights:
|
| 53 |
+
# - raw sharded safetensors: model.language_model.X
|
| 54 |
+
# - AutoModelForCausalLM.from_pretrained (text-only Qwen3_5ForCausalLM): model.X
|
| 55 |
+
# Strip whichever matches, longest first, so BOTH sources yield identical
|
| 56 |
+
# internal keys. Top-level weights like lm_head.weight have no prefix and are
|
| 57 |
+
# matched against the original `key` below.
|
| 58 |
+
new_key = key
|
| 59 |
+
for prefix in ("model.language_model.", "model."):
|
| 60 |
+
if new_key.startswith(prefix):
|
| 61 |
+
new_key = new_key[len(prefix) :]
|
| 62 |
+
break
|
| 63 |
+
|
| 64 |
+
# Rename top-level weights
|
| 65 |
+
if new_key == "embed_tokens.weight":
|
| 66 |
+
remapped["tok_embeddings.weight"] = tensor
|
| 67 |
+
continue
|
| 68 |
+
if key == "lm_head.weight":
|
| 69 |
+
remapped["output.weight"] = tensor
|
| 70 |
+
continue
|
| 71 |
+
# Final norm (model.language_model.norm.weight)
|
| 72 |
+
if new_key == "norm.weight":
|
| 73 |
+
remapped["norm.weight"] = tensor
|
| 74 |
+
continue
|
| 75 |
+
|
| 76 |
+
# Handle per-layer weights
|
| 77 |
+
if new_key.startswith("layers."):
|
| 78 |
+
parts = new_key.split(".")
|
| 79 |
+
layer_idx = int(parts[1])
|
| 80 |
+
layer_prefix = f"layers.{layer_idx}"
|
| 81 |
+
sub_key = ".".join(parts[2:])
|
| 82 |
+
|
| 83 |
+
# DeltaNet layers: keep ONLY the combined QKV weight. The split q/k/v_proj
|
| 84 |
+
# were dead — the op runs the fused QKV projection from the combined weight
|
| 85 |
+
# (it only read the splits in a fallback reached when qkv_proj_weight is None,
|
| 86 |
+
# which never happens for the 9B).
|
| 87 |
+
if sub_key == "linear_attn.in_proj_qkv.weight":
|
| 88 |
+
remapped[f"{layer_prefix}.linear_attn.qkv_proj.weight"] = tensor # [8192, 4096]
|
| 89 |
+
continue
|
| 90 |
+
|
| 91 |
+
if sub_key == "linear_attn.conv1d.weight":
|
| 92 |
+
conv = tensor # [8192, 1, 4]
|
| 93 |
+
q_conv = conv[:LINEAR_Q_DIM, :, :]
|
| 94 |
+
k_conv = conv[LINEAR_Q_DIM : LINEAR_Q_DIM + LINEAR_K_DIM, :, :]
|
| 95 |
+
v_conv = conv[LINEAR_Q_DIM + LINEAR_K_DIM :, :, :]
|
| 96 |
+
remapped[f"{layer_prefix}.linear_attn.q_conv.weight"] = q_conv
|
| 97 |
+
remapped[f"{layer_prefix}.linear_attn.k_conv.weight"] = k_conv
|
| 98 |
+
remapped[f"{layer_prefix}.linear_attn.v_conv.weight"] = v_conv
|
| 99 |
+
continue
|
| 100 |
+
|
| 101 |
+
# All other keys pass through unchanged
|
| 102 |
+
remapped[new_key] = tensor
|
| 103 |
+
continue
|
| 104 |
+
|
| 105 |
+
# Any remaining keys pass through
|
| 106 |
+
remapped[new_key] = tensor
|
| 107 |
+
|
| 108 |
+
return remapped
|
| 109 |
+
|
| 110 |
+
|
| 111 |
+
def load_qwen36_mtp_state_dict(model_path) -> Dict[str, torch.Tensor]:
|
| 112 |
+
"""Load only the checkpoint's small MTP subtree from sharded safetensors."""
|
| 113 |
+
from safetensors import safe_open
|
| 114 |
+
|
| 115 |
+
model_path = Path(model_path)
|
| 116 |
+
with open(model_path / "model.safetensors.index.json") as f:
|
| 117 |
+
weight_map = json.load(f)["weight_map"]
|
| 118 |
+
|
| 119 |
+
file_to_keys: Dict[str, list[str]] = {}
|
| 120 |
+
for key, filename in weight_map.items():
|
| 121 |
+
if key.startswith("mtp."):
|
| 122 |
+
file_to_keys.setdefault(filename, []).append(key)
|
| 123 |
+
|
| 124 |
+
mtp: Dict[str, torch.Tensor] = {}
|
| 125 |
+
for filename, keys in file_to_keys.items():
|
| 126 |
+
with safe_open(str(model_path / filename), framework="pt") as sf:
|
| 127 |
+
for key in keys:
|
| 128 |
+
mtp[key] = sf.get_tensor(key)
|
| 129 |
+
if not mtp:
|
| 130 |
+
raise ValueError(f"Qwen3.5 checkpoint at {model_path} has no mtp.* weights")
|
| 131 |
+
return mtp
|
| 132 |
+
|
| 133 |
+
|
| 134 |
+
def is_fp8_checkpoint(model_path) -> bool:
|
| 135 |
+
"""True when the checkpoint dir holds block-wise FP8 safetensors.
|
| 136 |
+
|
| 137 |
+
Detected by a ``*.weight_scale_inv`` entry in the safetensors index (the
|
| 138 |
+
per-block dequant scales that accompany float8_e4m3fn weights). Such a
|
| 139 |
+
checkpoint cannot be loaded via AutoModelForCausalLM here; use
|
| 140 |
+
``load_qwen36_state_dict_fp8`` instead.
|
| 141 |
+
"""
|
| 142 |
+
index_path = Path(model_path) / "model.safetensors.index.json"
|
| 143 |
+
if not index_path.is_file():
|
| 144 |
+
return False
|
| 145 |
+
try:
|
| 146 |
+
with open(index_path) as f:
|
| 147 |
+
weight_map = json.load(f)["weight_map"]
|
| 148 |
+
except (KeyError, ValueError, OSError):
|
| 149 |
+
return False
|
| 150 |
+
return any(k.endswith(".weight_scale_inv") for k in weight_map)
|
| 151 |
+
|
| 152 |
+
|
| 153 |
+
def load_qwen36_state_dict_fp8(model_path) -> Dict[str, torch.Tensor]:
|
| 154 |
+
"""Load Qwen3.5 FP8 weights: block-wise dequant + minimal key remap.
|
| 155 |
+
|
| 156 |
+
Produces the SAME internal key scheme as ``remap_qwen36_state_dict`` for the
|
| 157 |
+
shared/simple weights (``layers.N.mlp.*``, ``layers.N.self_attn.*``,
|
| 158 |
+
``input_layernorm`` / ``post_attention_layernorm``, ``tok_embeddings``,
|
| 159 |
+
``norm``, ``output``) so ``layer.py``'s substate extraction is unchanged.
|
| 160 |
+
|
| 161 |
+
The one deliberate difference vs the single-device remap: GDN ``linear_attn.*``
|
| 162 |
+
projections are kept RAW (fused ``in_proj_qkv``, fused ``conv1d``, plus
|
| 163 |
+
``in_proj_z`` / ``in_proj_a`` / ``in_proj_b`` / ``out_proj`` / ``A_log`` /
|
| 164 |
+
``dt_bias`` / ``norm.weight``) — NOT split or renamed — so the tensor-parallel
|
| 165 |
+
GDN weight-prep helpers (prepare_gdn_qkv / prepare_conv_taps) can reorder and
|
| 166 |
+
shard them per device. The TP module loaders branch on this raw layout.
|
| 167 |
+
"""
|
| 168 |
+
from safetensors import safe_open
|
| 169 |
+
|
| 170 |
+
from models.demos.blackhole.qwen36.tt.tp_common import dequant_fp8_block
|
| 171 |
+
|
| 172 |
+
model_path = Path(model_path)
|
| 173 |
+
index_path = model_path / "model.safetensors.index.json"
|
| 174 |
+
with open(index_path) as f:
|
| 175 |
+
weight_map = json.load(f)["weight_map"]
|
| 176 |
+
|
| 177 |
+
file_to_keys: Dict[str, list] = {}
|
| 178 |
+
for key, filename in weight_map.items():
|
| 179 |
+
file_to_keys.setdefault(filename, []).append(key)
|
| 180 |
+
|
| 181 |
+
raw: Dict[str, torch.Tensor] = {}
|
| 182 |
+
for filename, keys in file_to_keys.items():
|
| 183 |
+
with safe_open(str(model_path / filename), framework="pt") as sf:
|
| 184 |
+
present = set(sf.keys())
|
| 185 |
+
for key in keys:
|
| 186 |
+
if key in present:
|
| 187 |
+
raw[key] = sf.get_tensor(key)
|
| 188 |
+
|
| 189 |
+
# Dequantize FP8 (skip the scale tensors themselves)
|
| 190 |
+
dequantized: Dict[str, torch.Tensor] = {}
|
| 191 |
+
for key, tensor in raw.items():
|
| 192 |
+
if key.endswith(".weight_scale_inv"):
|
| 193 |
+
continue
|
| 194 |
+
if tensor.dtype == torch.float8_e4m3fn:
|
| 195 |
+
scale_key = key + "_scale_inv"
|
| 196 |
+
dequantized[key] = (
|
| 197 |
+
dequant_fp8_block(tensor, raw[scale_key]) if scale_key in raw else tensor.to(torch.bfloat16)
|
| 198 |
+
)
|
| 199 |
+
else:
|
| 200 |
+
dequantized[key] = tensor
|
| 201 |
+
|
| 202 |
+
state_dict: Dict[str, torch.Tensor] = {}
|
| 203 |
+
for key, tensor in dequantized.items():
|
| 204 |
+
if "visual" in key or key.startswith("mtp"):
|
| 205 |
+
continue
|
| 206 |
+
short = key
|
| 207 |
+
for prefix in ("model.language_model.", "model."):
|
| 208 |
+
if short.startswith(prefix):
|
| 209 |
+
short = short[len(prefix) :]
|
| 210 |
+
break
|
| 211 |
+
if "embed_tokens" in short:
|
| 212 |
+
state_dict["tok_embeddings.weight"] = tensor
|
| 213 |
+
elif key == "lm_head.weight" or short == "lm_head.weight":
|
| 214 |
+
state_dict["output.weight"] = tensor
|
| 215 |
+
else:
|
| 216 |
+
# Everything else (layers.N.mlp.*, self_attn.*, linear_attn.* RAW,
|
| 217 |
+
# input_layernorm/post_attention_layernorm, norm) passes through.
|
| 218 |
+
state_dict[short] = tensor
|
| 219 |
+
|
| 220 |
+
return state_dict
|
checkpoint/quantization-error.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint/tensors/00000.tensorbin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0d4f553fe416591e5d9dfdf45150f82b4d88b4fea590983b65fe126bf602ee5a
|
| 3 |
+
size 1080689000
|
checkpoint/tensors/00001.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:38bb1a27e20c64787cd377c3f18bf2117aa4ed8251d247aedb146bb4833b0573
|
| 3 |
+
size 2034237544
|
checkpoint/tensors/00002.tensorbin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b55492af5a288f52ebe592e39e2e75f6949706869b679621b41597619f9a7512
|
| 3 |
+
size 28311912
|
checkpoint/tensors/00003.tensorbin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:817739482fde7bf45eab7b3d68831fb27490390928c60966d48084fae7127b13
|
| 3 |
+
size 53477736
|
checkpoint/tensors/00004.tensorbin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d0552d5ca5f484aa8fce00132cef93c03ebfdb2c0cf2bb79c0e48ca2d787f971
|
| 3 |
+
size 53477736
|
checkpoint/tensors/00005.tensorbin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b5a8cbeb04c20ae79baf8ccd161ec5a05403ee15c6974e657a883f98769b0888
|
| 3 |
+
size 28311912
|
checkpoint/tensors/00006.tensorbin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2663e15be67b452fcb6836f53efb57b1ebf7bc078889ef94c3d10973164647b6
|
| 3 |
+
size 53477736
|
checkpoint/tensors/00007.tensorbin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8d1d32db0f18d850c2af69f9972a86dc53b789f604cbb38e934078ba4e98e833
|
| 3 |
+
size 53477736
|
checkpoint/tensors/00008.tensorbin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dbceb9cee219c05d5331cca9d3fe128834dba6ac763db5216ea0cfc58d4c3e11
|
| 3 |
+
size 28311912
|
checkpoint/tensors/00009.tensorbin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1adcc930caac883061d6b927717fbc139896ee40754bc7970616a4dfc85666cc
|
| 3 |
+
size 28311912
|
checkpoint/tensors/00010.tensorbin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1f0b6775ab6d1429b849a7547fa1e83d30d26a179fea984875d9845838208052
|
| 3 |
+
size 28311912
|
checkpoint/tensors/00011.tensorbin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:752d18b915e7e3042dccce83624c75fa99879d390689a91b5fcdb72ff51ba8b3
|
| 3 |
+
size 28311912
|
checkpoint/tensors/00012.tensorbin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b5a07da9f758a7bd30153ee609e82be88ec4f6ec82c5693c72a66dd216315269
|
| 3 |
+
size 28311912
|
checkpoint/tensors/00013.tensorbin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e019672df4f09a8f8cbaa8851a9c0e6405dc243f8fd76f9ae64599bc0f6cf6ce
|
| 3 |
+
size 53477736
|
checkpoint/tensors/00014.tensorbin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:470eff8f523155457cdeb861853b0ef8c0909399f3922f70424dd60f35d39cf0
|
| 3 |
+
size 28311912
|
checkpoint/tensors/00015.tensorbin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0078928a85361a6b3b372725b25e9edb2332effce2dae748ad394819d624caac
|
| 3 |
+
size 28311912
|
checkpoint/tensors/00016.tensorbin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:88c0a18c5c2ea0c5bbaa99d9d3dd782b28cc184a3b4528adcbf66a0f27d82709
|
| 3 |
+
size 28311912
|
checkpoint/tensors/00017.tensorbin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:161db47ca7a7747ec28c78a82b0b57ce46478ec7cfe2cd9f1f179fd0e44041b1
|
| 3 |
+
size 53477736
|
checkpoint/tensors/00018.tensorbin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4c2de1c8bc2518ffaceb0b3ed5c0a1ff1de0ffb3cd8b3d684635d01cafd96fd3
|
| 3 |
+
size 28311912
|
checkpoint/tensors/00019.tensorbin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6710c4b861036d61791c017dd22e7ec747abe741412e7f31d56b56014dfc2746
|
| 3 |
+
size 28311912
|
checkpoint/tensors/00020.tensorbin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:37726ca2810b3114267b599c1fff6dd17c18b534be98349851a74a0031511fd9
|
| 3 |
+
size 28311912
|
checkpoint/tensors/00021.tensorbin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:620b7f5bf138cef6f091b94e7bb4260e4e9693abcfc9257e6a7788af062a7ae0
|
| 3 |
+
size 28311912
|
checkpoint/tensors/00022.tensorbin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:70cd2c36ae9b0c518ec4ed9ccba038448c8b07a77fc40a61dba1b4e47ed403b6
|
| 3 |
+
size 28311912
|