Hugging Face's logo

frankmorales2020
/
mixtral-8x7b-fp8-topo2026

Safetensors
mixtral
continual-learning
catastrophic-forgetting
topological-ai
TOPO-2026
compressed-tensors
Model card Files
xet
 Community
frankmorales2020 commited on
Commit
88db8d8
·
verified ·
1 Parent(s): 5f3706a

Mixtral-8x7B FP8 TOPO-2026 | Embeddings BF16 | MoE gates BF16 | LLMCompressor

Browse files
Files changed (18) hide show
  1. config.json +113 -0
  2. generation_config.json +6 -0
  3. model-00001-of-00010.safetensors +3 -0
  4. model-00002-of-00010.safetensors +3 -0
  5. model-00003-of-00010.safetensors +3 -0
  6. model-00004-of-00010.safetensors +3 -0
  7. model-00005-of-00010.safetensors +3 -0
  8. model-00006-of-00010.safetensors +3 -0
  9. model-00007-of-00010.safetensors +3 -0
  10. model-00008-of-00010.safetensors +3 -0
  11. model-00009-of-00010.safetensors +3 -0
  12. model-00010-of-00010.safetensors +3 -0
  13. model.safetensors.index.json +0 -0
  14. recipe.yaml +7 -0
  15. special_tokens_map.json +24 -0
  16. tokenizer.json +0 -0
  17. tokenizer.model +3 -0
  18. tokenizer_config.json +44 -0
config.json ADDED
@@ -0,0 +1,113 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "MixtralForCausalLM"
4
+ ],
5
+ "attention_dropout": 0.0,
6
+ "bos_token_id": 1,
7
+ "dtype": "bfloat16",
8
+ "eos_token_id": 2,
9
+ "head_dim": null,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 4096,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 14336,
14
+ "max_position_embeddings": 32768,
15
+ "model_type": "mixtral",
16
+ "num_attention_heads": 32,
17
+ "num_experts_per_tok": 2,
18
+ "num_hidden_layers": 32,
19
+ "num_key_value_heads": 8,
20
+ "num_local_experts": 8,
21
+ "output_router_logits": false,
22
+ "quantization_config": {
23
+ "config_groups": {
24
+ "group_0": {
25
+ "format": "float-quantized",
26
+ "input_activations": {
27
+ "actorder": null,
28
+ "block_structure": null,
29
+ "dynamic": false,
30
+ "group_size": null,
31
+ "num_bits": 8,
32
+ "observer": "memoryless_minmax",
33
+ "observer_kwargs": {},
34
+ "scale_dtype": null,
35
+ "strategy": "tensor",
36
+ "symmetric": true,
37
+ "type": "float",
38
+ "zp_dtype": null
39
+ },
40
+ "output_activations": null,
41
+ "targets": [
42
+ "Linear"
43
+ ],
44
+ "weights": {
45
+ "actorder": null,
46
+ "block_structure": null,
47
+ "dynamic": false,
48
+ "group_size": null,
49
+ "num_bits": 8,
50
+ "observer": "memoryless_minmax",
51
+ "observer_kwargs": {},
52
+ "scale_dtype": null,
53
+ "strategy": "tensor",
54
+ "symmetric": true,
55
+ "type": "float",
56
+ "zp_dtype": null
57
+ }
58
+ }
59
+ },
60
+ "format": "float-quantized",
61
+ "global_compression_ratio": null,
62
+ "ignore": [
63
+ "model.layers.0.block_sparse_moe.gate",
64
+ "model.layers.1.block_sparse_moe.gate",
65
+ "model.layers.2.block_sparse_moe.gate",
66
+ "model.layers.3.block_sparse_moe.gate",
67
+ "model.layers.4.block_sparse_moe.gate",
68
+ "model.layers.5.block_sparse_moe.gate",
69
+ "model.layers.6.block_sparse_moe.gate",
70
+ "model.layers.7.block_sparse_moe.gate",
71
+ "model.layers.8.block_sparse_moe.gate",
72
+ "model.layers.9.block_sparse_moe.gate",
73
+ "model.layers.10.block_sparse_moe.gate",
74
+ "model.layers.11.block_sparse_moe.gate",
75
+ "model.layers.12.block_sparse_moe.gate",
76
+ "model.layers.13.block_sparse_moe.gate",
77
+ "model.layers.14.block_sparse_moe.gate",
78
+ "model.layers.15.block_sparse_moe.gate",
79
+ "model.layers.16.block_sparse_moe.gate",
80
+ "model.layers.17.block_sparse_moe.gate",
81
+ "model.layers.18.block_sparse_moe.gate",
82
+ "model.layers.19.block_sparse_moe.gate",
83
+ "model.layers.20.block_sparse_moe.gate",
84
+ "model.layers.21.block_sparse_moe.gate",
85
+ "model.layers.22.block_sparse_moe.gate",
86
+ "model.layers.23.block_sparse_moe.gate",
87
+ "model.layers.24.block_sparse_moe.gate",
88
+ "model.layers.25.block_sparse_moe.gate",
89
+ "model.layers.26.block_sparse_moe.gate",
90
+ "model.layers.27.block_sparse_moe.gate",
91
+ "model.layers.28.block_sparse_moe.gate",
92
+ "model.layers.29.block_sparse_moe.gate",
93
+ "model.layers.30.block_sparse_moe.gate",
94
+ "model.layers.31.block_sparse_moe.gate",
95
+ "lm_head"
96
+ ],
97
+ "kv_cache_scheme": null,
98
+ "quant_method": "compressed-tensors",
99
+ "quantization_status": "compressed",
100
+ "sparsity_config": {},
101
+ "transform_config": {},
102
+ "version": "0.14.0.1"
103
+ },
104
+ "rms_norm_eps": 1e-05,
105
+ "rope_theta": 1000000.0,
106
+ "router_aux_loss_coef": 0.02,
107
+ "router_jitter_noise": 0.0,
108
+ "sliding_window": null,
109
+ "tie_word_embeddings": false,
110
+ "transformers_version": "4.57.6",
111
+ "use_cache": true,
112
+ "vocab_size": 32000
113
+ }
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "transformers_version": "4.57.6"
6
+ }
model-00001-of-00010.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a6ce9a535c08c81a41c220991d984414ebf427a4f8a6fbf95d819c311f468768
3
+ size 4951723964
model-00002-of-00010.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:74df03a9e421c24f58a7fbaed5bf76cfe9a4740af47a163d501e6168a9fd95c5
3
+ size 4999891908
model-00003-of-00010.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b64a244cdc142c4c12b0a9f7386aee7af3543e599eb7086480ab62d13da8ec31
3
+ size 4983198264
model-00004-of-00010.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81735d64ecb4426fbc839c05fce3d9ad270c22bdcb1058a9c1aa327d1bf66523
3
+ size 4999892156
model-00005-of-00010.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a5b0fdd9b8fee3175d4d807a217bd20f7b5ef6f93f8e8a3bcd61b81c860d2ab6
3
+ size 4999908932
model-00006-of-00010.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:54b23d4cc0ed351b02c5fcefa64ed94da4251c54b1b3bf7e8cc7ae9d38fee77f
3
+ size 4983181632
model-00007-of-00010.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:61f95aefc74b2a8f6fe74b7acd500f863f52b4a69d1104853ad70350f84c83d1
3
+ size 4999892284
model-00008-of-00010.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:715b6cbbcb6d96cc5705e3856c20dcfede32d66a32ab2528205d10a0230e95dd
3
+ size 4983198472
model-00009-of-00010.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:020fe894a506e7279e202251fe978a62df46c8c2d062db980d13da8ae5fbdcaf
3
+ size 4999892236
model-00010-of-00010.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c93b1959424b686f7236e60aa8f4cf03052ced7d258a89d95754d75d0ec8cdac
3
+ size 2065815152
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
recipe.yaml ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ default_stage:
2
+ default_modifiers:
3
+ QuantizationModifier:
4
+ targets: [Linear]
5
+ ignore: [lm_head, 're:.*gate.*', 're:.*embed.*']
6
+ scheme: FP8
7
+ bypass_divisibility_checks: false
special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "</s>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
3
+ size 493443
tokenizer_config.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": null,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ }
30
+ },
31
+ "additional_special_tokens": [],
32
+ "bos_token": "<s>",
33
+ "clean_up_tokenization_spaces": false,
34
+ "eos_token": "</s>",
35
+ "extra_special_tokens": {},
36
+ "legacy": true,
37
+ "model_max_length": 1000000000000000019884624838656,
38
+ "pad_token": "</s>",
39
+ "sp_model_kwargs": {},
40
+ "spaces_between_special_tokens": false,
41
+ "tokenizer_class": "LlamaTokenizer",
42
+ "unk_token": "<unk>",
43
+ "use_default_system_prompt": false
44
+ }