Hugging Face's logo

TartarusXXX
/
whisper-lora-georgian-config-1

Transformers
Safetensors
unsloth
Model card Files
xet
 Community
EmreAkgul commited on
Commit
7fc07d0
·
verified ·
1 Parent(s): e13ff06

Training in progress, step 5000

Browse files
Files changed (4) hide show
  1. adapter_config.json +208 -0
  2. adapter_model.safetensors +3 -0
  3. preprocessor_config.json +15 -0
  4. training_args.bin +3 -0
adapter_config.json ADDED
@@ -0,0 +1,208 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": {
4
+ "base_model_class": "WhisperForConditionalGeneration",
5
+ "parent_library": "transformers.models.whisper.modeling_whisper"
6
+ },
7
+ "base_model_name_or_path": "unsloth/whisper-large-v3",
8
+ "bias": "none",
9
+ "corda_config": null,
10
+ "eva_config": null,
11
+ "exclude_modules": null,
12
+ "fan_in_fan_out": false,
13
+ "inference_mode": true,
14
+ "init_lora_weights": true,
15
+ "layer_replication": null,
16
+ "layers_pattern": null,
17
+ "layers_to_transform": null,
18
+ "loftq_config": {},
19
+ "lora_alpha": 64,
20
+ "lora_bias": false,
21
+ "lora_dropout": 0.03,
22
+ "megatron_config": null,
23
+ "megatron_core": "megatron.core",
24
+ "modules_to_save": [
25
+ "model.encoder.layers.0.self_attn_layer_norm",
26
+ "model.encoder.layers.0.final_layer_norm",
27
+ "model.encoder.layers.1.self_attn_layer_norm",
28
+ "model.encoder.layers.1.final_layer_norm",
29
+ "model.encoder.layers.2.self_attn_layer_norm",
30
+ "model.encoder.layers.2.final_layer_norm",
31
+ "model.encoder.layers.3.self_attn_layer_norm",
32
+ "model.encoder.layers.3.final_layer_norm",
33
+ "model.encoder.layers.4.self_attn_layer_norm",
34
+ "model.encoder.layers.4.final_layer_norm",
35
+ "model.encoder.layers.5.self_attn_layer_norm",
36
+ "model.encoder.layers.5.final_layer_norm",
37
+ "model.encoder.layers.6.self_attn_layer_norm",
38
+ "model.encoder.layers.6.final_layer_norm",
39
+ "model.encoder.layers.7.self_attn_layer_norm",
40
+ "model.encoder.layers.7.final_layer_norm",
41
+ "model.encoder.layers.8.self_attn_layer_norm",
42
+ "model.encoder.layers.8.final_layer_norm",
43
+ "model.encoder.layers.9.self_attn_layer_norm",
44
+ "model.encoder.layers.9.final_layer_norm",
45
+ "model.encoder.layers.10.self_attn_layer_norm",
46
+ "model.encoder.layers.10.final_layer_norm",
47
+ "model.encoder.layers.11.self_attn_layer_norm",
48
+ "model.encoder.layers.11.final_layer_norm",
49
+ "model.encoder.layers.12.self_attn_layer_norm",
50
+ "model.encoder.layers.12.final_layer_norm",
51
+ "model.encoder.layers.13.self_attn_layer_norm",
52
+ "model.encoder.layers.13.final_layer_norm",
53
+ "model.encoder.layers.14.self_attn_layer_norm",
54
+ "model.encoder.layers.14.final_layer_norm",
55
+ "model.encoder.layers.15.self_attn_layer_norm",
56
+ "model.encoder.layers.15.final_layer_norm",
57
+ "model.encoder.layers.16.self_attn_layer_norm",
58
+ "model.encoder.layers.16.final_layer_norm",
59
+ "model.encoder.layers.17.self_attn_layer_norm",
60
+ "model.encoder.layers.17.final_layer_norm",
61
+ "model.encoder.layers.18.self_attn_layer_norm",
62
+ "model.encoder.layers.18.final_layer_norm",
63
+ "model.encoder.layers.19.self_attn_layer_norm",
64
+ "model.encoder.layers.19.final_layer_norm",
65
+ "model.encoder.layers.20.self_attn_layer_norm",
66
+ "model.encoder.layers.20.final_layer_norm",
67
+ "model.encoder.layers.21.self_attn_layer_norm",
68
+ "model.encoder.layers.21.final_layer_norm",
69
+ "model.encoder.layers.22.self_attn_layer_norm",
70
+ "model.encoder.layers.22.final_layer_norm",
71
+ "model.encoder.layers.23.self_attn_layer_norm",
72
+ "model.encoder.layers.23.final_layer_norm",
73
+ "model.encoder.layers.24.self_attn_layer_norm",
74
+ "model.encoder.layers.24.final_layer_norm",
75
+ "model.encoder.layers.25.self_attn_layer_norm",
76
+ "model.encoder.layers.25.final_layer_norm",
77
+ "model.encoder.layers.26.self_attn_layer_norm",
78
+ "model.encoder.layers.26.final_layer_norm",
79
+ "model.encoder.layers.27.self_attn_layer_norm",
80
+ "model.encoder.layers.27.final_layer_norm",
81
+ "model.encoder.layers.28.self_attn_layer_norm",
82
+ "model.encoder.layers.28.final_layer_norm",
83
+ "model.encoder.layers.29.self_attn_layer_norm",
84
+ "model.encoder.layers.29.final_layer_norm",
85
+ "model.encoder.layers.30.self_attn_layer_norm",
86
+ "model.encoder.layers.30.final_layer_norm",
87
+ "model.encoder.layers.31.self_attn_layer_norm",
88
+ "model.encoder.layers.31.final_layer_norm",
89
+ "model.encoder.layer_norm",
90
+ "model.decoder.layers.0.self_attn_layer_norm",
91
+ "model.decoder.layers.0.encoder_attn_layer_norm",
92
+ "model.decoder.layers.0.final_layer_norm",
93
+ "model.decoder.layers.1.self_attn_layer_norm",
94
+ "model.decoder.layers.1.encoder_attn_layer_norm",
95
+ "model.decoder.layers.1.final_layer_norm",
96
+ "model.decoder.layers.2.self_attn_layer_norm",
97
+ "model.decoder.layers.2.encoder_attn_layer_norm",
98
+ "model.decoder.layers.2.final_layer_norm",
99
+ "model.decoder.layers.3.self_attn_layer_norm",
100
+ "model.decoder.layers.3.encoder_attn_layer_norm",
101
+ "model.decoder.layers.3.final_layer_norm",
102
+ "model.decoder.layers.4.self_attn_layer_norm",
103
+ "model.decoder.layers.4.encoder_attn_layer_norm",
104
+ "model.decoder.layers.4.final_layer_norm",
105
+ "model.decoder.layers.5.self_attn_layer_norm",
106
+ "model.decoder.layers.5.encoder_attn_layer_norm",
107
+ "model.decoder.layers.5.final_layer_norm",
108
+ "model.decoder.layers.6.self_attn_layer_norm",
109
+ "model.decoder.layers.6.encoder_attn_layer_norm",
110
+ "model.decoder.layers.6.final_layer_norm",
111
+ "model.decoder.layers.7.self_attn_layer_norm",
112
+ "model.decoder.layers.7.encoder_attn_layer_norm",
113
+ "model.decoder.layers.7.final_layer_norm",
114
+ "model.decoder.layers.8.self_attn_layer_norm",
115
+ "model.decoder.layers.8.encoder_attn_layer_norm",
116
+ "model.decoder.layers.8.final_layer_norm",
117
+ "model.decoder.layers.9.self_attn_layer_norm",
118
+ "model.decoder.layers.9.encoder_attn_layer_norm",
119
+ "model.decoder.layers.9.final_layer_norm",
120
+ "model.decoder.layers.10.self_attn_layer_norm",
121
+ "model.decoder.layers.10.encoder_attn_layer_norm",
122
+ "model.decoder.layers.10.final_layer_norm",
123
+ "model.decoder.layers.11.self_attn_layer_norm",
124
+ "model.decoder.layers.11.encoder_attn_layer_norm",
125
+ "model.decoder.layers.11.final_layer_norm",
126
+ "model.decoder.layers.12.self_attn_layer_norm",
127
+ "model.decoder.layers.12.encoder_attn_layer_norm",
128
+ "model.decoder.layers.12.final_layer_norm",
129
+ "model.decoder.layers.13.self_attn_layer_norm",
130
+ "model.decoder.layers.13.encoder_attn_layer_norm",
131
+ "model.decoder.layers.13.final_layer_norm",
132
+ "model.decoder.layers.14.self_attn_layer_norm",
133
+ "model.decoder.layers.14.encoder_attn_layer_norm",
134
+ "model.decoder.layers.14.final_layer_norm",
135
+ "model.decoder.layers.15.self_attn_layer_norm",
136
+ "model.decoder.layers.15.encoder_attn_layer_norm",
137
+ "model.decoder.layers.15.final_layer_norm",
138
+ "model.decoder.layers.16.self_attn_layer_norm",
139
+ "model.decoder.layers.16.encoder_attn_layer_norm",
140
+ "model.decoder.layers.16.final_layer_norm",
141
+ "model.decoder.layers.17.self_attn_layer_norm",
142
+ "model.decoder.layers.17.encoder_attn_layer_norm",
143
+ "model.decoder.layers.17.final_layer_norm",
144
+ "model.decoder.layers.18.self_attn_layer_norm",
145
+ "model.decoder.layers.18.encoder_attn_layer_norm",
146
+ "model.decoder.layers.18.final_layer_norm",
147
+ "model.decoder.layers.19.self_attn_layer_norm",
148
+ "model.decoder.layers.19.encoder_attn_layer_norm",
149
+ "model.decoder.layers.19.final_layer_norm",
150
+ "model.decoder.layers.20.self_attn_layer_norm",
151
+ "model.decoder.layers.20.encoder_attn_layer_norm",
152
+ "model.decoder.layers.20.final_layer_norm",
153
+ "model.decoder.layers.21.self_attn_layer_norm",
154
+ "model.decoder.layers.21.encoder_attn_layer_norm",
155
+ "model.decoder.layers.21.final_layer_norm",
156
+ "model.decoder.layers.22.self_attn_layer_norm",
157
+ "model.decoder.layers.22.encoder_attn_layer_norm",
158
+ "model.decoder.layers.22.final_layer_norm",
159
+ "model.decoder.layers.23.self_attn_layer_norm",
160
+ "model.decoder.layers.23.encoder_attn_layer_norm",
161
+ "model.decoder.layers.23.final_layer_norm",
162
+ "model.decoder.layers.24.self_attn_layer_norm",
163
+ "model.decoder.layers.24.encoder_attn_layer_norm",
164
+ "model.decoder.layers.24.final_layer_norm",
165
+ "model.decoder.layers.25.self_attn_layer_norm",
166
+ "model.decoder.layers.25.encoder_attn_layer_norm",
167
+ "model.decoder.layers.25.final_layer_norm",
168
+ "model.decoder.layers.26.self_attn_layer_norm",
169
+ "model.decoder.layers.26.encoder_attn_layer_norm",
170
+ "model.decoder.layers.26.final_layer_norm",
171
+ "model.decoder.layers.27.self_attn_layer_norm",
172
+ "model.decoder.layers.27.encoder_attn_layer_norm",
173
+ "model.decoder.layers.27.final_layer_norm",
174
+ "model.decoder.layers.28.self_attn_layer_norm",
175
+ "model.decoder.layers.28.encoder_attn_layer_norm",
176
+ "model.decoder.layers.28.final_layer_norm",
177
+ "model.decoder.layers.29.self_attn_layer_norm",
178
+ "model.decoder.layers.29.encoder_attn_layer_norm",
179
+ "model.decoder.layers.29.final_layer_norm",
180
+ "model.decoder.layers.30.self_attn_layer_norm",
181
+ "model.decoder.layers.30.encoder_attn_layer_norm",
182
+ "model.decoder.layers.30.final_layer_norm",
183
+ "model.decoder.layers.31.self_attn_layer_norm",
184
+ "model.decoder.layers.31.encoder_attn_layer_norm",
185
+ "model.decoder.layers.31.final_layer_norm",
186
+ "model.decoder.layer_norm"
187
+ ],
188
+ "peft_type": "LORA",
189
+ "qalora_group_size": 16,
190
+ "r": 64,
191
+ "rank_pattern": {},
192
+ "revision": null,
193
+ "target_modules": [
194
+ "gate_proj",
195
+ "q_proj",
196
+ "v_proj",
197
+ "down_proj",
198
+ "up_proj",
199
+ "o_proj",
200
+ "k_proj"
201
+ ],
202
+ "target_parameters": null,
203
+ "task_type": null,
204
+ "trainable_token_indices": null,
205
+ "use_dora": false,
206
+ "use_qalora": false,
207
+ "use_rslora": false
208
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b2019612743b11f38edbc5f57cccb8af5bc6c17e868cf3f6511075c010acadbc
3
+ size 189700928
preprocessor_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "chunk_length": 30,
3
+ "dither": 0.0,
4
+ "feature_extractor_type": "WhisperFeatureExtractor",
5
+ "feature_size": 128,
6
+ "hop_length": 160,
7
+ "n_fft": 400,
8
+ "n_samples": 480000,
9
+ "nb_max_frames": 3000,
10
+ "padding_side": "left",
11
+ "padding_value": 0.0,
12
+ "processor_class": "WhisperProcessor",
13
+ "return_attention_mask": false,
14
+ "sampling_rate": 16000
15
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d8defff7cf4a8249517c560b6ab9bd3fc84b0f9bda399791c2ed26182bc77f73
3
+ size 6033