otozz commited on
Commit
eb52ba4
·
verified ·
1 Parent(s): 033e4ab

Upload folder using huggingface_hub

Browse files
._config.json ADDED
Binary file (4.1 kB). View file
 
._generation_config.json ADDED
Binary file (4.1 kB). View file
 
._model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70dd60c5e38bdd023df1e968059e8a092b061648449e42c673d9d32d1f8a4901
3
+ size 4096
._optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70dd60c5e38bdd023df1e968059e8a092b061648449e42c673d9d32d1f8a4901
3
+ size 4096
._preprocessor_config.json ADDED
Binary file (4.1 kB). View file
 
._rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70dd60c5e38bdd023df1e968059e8a092b061648449e42c673d9d32d1f8a4901
3
+ size 4096
._scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70dd60c5e38bdd023df1e968059e8a092b061648449e42c673d9d32d1f8a4901
3
+ size 4096
._trainer_state.json ADDED
Binary file (4.1 kB). View file
 
._training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70dd60c5e38bdd023df1e968059e8a092b061648449e42c673d9d32d1f8a4901
3
+ size 4096
config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "openai/whisper-small",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "gelu",
5
+ "apply_spec_augment": false,
6
+ "architectures": [
7
+ "WhisperForConditionalGeneration"
8
+ ],
9
+ "attention_dropout": 0.0,
10
+ "begin_suppress_tokens": [
11
+ 220,
12
+ 50257
13
+ ],
14
+ "bos_token_id": 50257,
15
+ "classifier_proj_size": 256,
16
+ "d_model": 768,
17
+ "decoder_attention_heads": 12,
18
+ "decoder_ffn_dim": 3072,
19
+ "decoder_layerdrop": 0.0,
20
+ "decoder_layers": 12,
21
+ "decoder_start_token_id": 50258,
22
+ "dropout": 0.0,
23
+ "encoder_attention_heads": 12,
24
+ "encoder_ffn_dim": 3072,
25
+ "encoder_layerdrop": 0.0,
26
+ "encoder_layers": 12,
27
+ "eos_token_id": 50257,
28
+ "forced_decoder_ids": null,
29
+ "init_std": 0.02,
30
+ "is_encoder_decoder": true,
31
+ "mask_feature_length": 10,
32
+ "mask_feature_min_masks": 0,
33
+ "mask_feature_prob": 0.0,
34
+ "mask_time_length": 10,
35
+ "mask_time_min_masks": 2,
36
+ "mask_time_prob": 0.05,
37
+ "max_length": 512,
38
+ "max_source_positions": 1500,
39
+ "max_target_positions": 448,
40
+ "median_filter_width": 7,
41
+ "model_type": "whisper",
42
+ "num_hidden_layers": 12,
43
+ "num_mel_bins": 80,
44
+ "pad_token_id": 50257,
45
+ "scale_embedding": false,
46
+ "suppress_tokens": [],
47
+ "torch_dtype": "float32",
48
+ "transformers_version": "4.40.1",
49
+ "use_cache": true,
50
+ "use_weighted_layer_sum": false,
51
+ "vocab_size": 51865
52
+ }
generation_config.json ADDED
@@ -0,0 +1,265 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alignment_heads": [
3
+ [
4
+ 5,
5
+ 3
6
+ ],
7
+ [
8
+ 5,
9
+ 9
10
+ ],
11
+ [
12
+ 8,
13
+ 0
14
+ ],
15
+ [
16
+ 8,
17
+ 4
18
+ ],
19
+ [
20
+ 8,
21
+ 7
22
+ ],
23
+ [
24
+ 8,
25
+ 8
26
+ ],
27
+ [
28
+ 9,
29
+ 0
30
+ ],
31
+ [
32
+ 9,
33
+ 7
34
+ ],
35
+ [
36
+ 9,
37
+ 9
38
+ ],
39
+ [
40
+ 10,
41
+ 5
42
+ ]
43
+ ],
44
+ "begin_suppress_tokens": [
45
+ 220,
46
+ 50257
47
+ ],
48
+ "bos_token_id": 50257,
49
+ "decoder_start_token_id": 50258,
50
+ "eos_token_id": 50257,
51
+ "forced_decoder_ids": [
52
+ [
53
+ 1,
54
+ null
55
+ ],
56
+ [
57
+ 2,
58
+ 50359
59
+ ]
60
+ ],
61
+ "is_multilingual": true,
62
+ "lang_to_id": {
63
+ "<|af|>": 50327,
64
+ "<|am|>": 50334,
65
+ "<|ar|>": 50272,
66
+ "<|as|>": 50350,
67
+ "<|az|>": 50304,
68
+ "<|ba|>": 50355,
69
+ "<|be|>": 50330,
70
+ "<|bg|>": 50292,
71
+ "<|bn|>": 50302,
72
+ "<|bo|>": 50347,
73
+ "<|br|>": 50309,
74
+ "<|bs|>": 50315,
75
+ "<|ca|>": 50270,
76
+ "<|cs|>": 50283,
77
+ "<|cy|>": 50297,
78
+ "<|da|>": 50285,
79
+ "<|de|>": 50261,
80
+ "<|el|>": 50281,
81
+ "<|en|>": 50259,
82
+ "<|es|>": 50262,
83
+ "<|et|>": 50307,
84
+ "<|eu|>": 50310,
85
+ "<|fa|>": 50300,
86
+ "<|fi|>": 50277,
87
+ "<|fo|>": 50338,
88
+ "<|fr|>": 50265,
89
+ "<|gl|>": 50319,
90
+ "<|gu|>": 50333,
91
+ "<|haw|>": 50352,
92
+ "<|ha|>": 50354,
93
+ "<|he|>": 50279,
94
+ "<|hi|>": 50276,
95
+ "<|hr|>": 50291,
96
+ "<|ht|>": 50339,
97
+ "<|hu|>": 50286,
98
+ "<|hy|>": 50312,
99
+ "<|id|>": 50275,
100
+ "<|is|>": 50311,
101
+ "<|it|>": 50274,
102
+ "<|ja|>": 50266,
103
+ "<|jw|>": 50356,
104
+ "<|ka|>": 50329,
105
+ "<|kk|>": 50316,
106
+ "<|km|>": 50323,
107
+ "<|kn|>": 50306,
108
+ "<|ko|>": 50264,
109
+ "<|la|>": 50294,
110
+ "<|lb|>": 50345,
111
+ "<|ln|>": 50353,
112
+ "<|lo|>": 50336,
113
+ "<|lt|>": 50293,
114
+ "<|lv|>": 50301,
115
+ "<|mg|>": 50349,
116
+ "<|mi|>": 50295,
117
+ "<|mk|>": 50308,
118
+ "<|ml|>": 50296,
119
+ "<|mn|>": 50314,
120
+ "<|mr|>": 50320,
121
+ "<|ms|>": 50282,
122
+ "<|mt|>": 50343,
123
+ "<|my|>": 50346,
124
+ "<|ne|>": 50313,
125
+ "<|nl|>": 50271,
126
+ "<|nn|>": 50342,
127
+ "<|no|>": 50288,
128
+ "<|oc|>": 50328,
129
+ "<|pa|>": 50321,
130
+ "<|pl|>": 50269,
131
+ "<|ps|>": 50340,
132
+ "<|pt|>": 50267,
133
+ "<|ro|>": 50284,
134
+ "<|ru|>": 50263,
135
+ "<|sa|>": 50344,
136
+ "<|sd|>": 50332,
137
+ "<|si|>": 50322,
138
+ "<|sk|>": 50298,
139
+ "<|sl|>": 50305,
140
+ "<|sn|>": 50324,
141
+ "<|so|>": 50326,
142
+ "<|sq|>": 50317,
143
+ "<|sr|>": 50303,
144
+ "<|su|>": 50357,
145
+ "<|sv|>": 50273,
146
+ "<|sw|>": 50318,
147
+ "<|ta|>": 50287,
148
+ "<|te|>": 50299,
149
+ "<|tg|>": 50331,
150
+ "<|th|>": 50289,
151
+ "<|tk|>": 50341,
152
+ "<|tl|>": 50348,
153
+ "<|tr|>": 50268,
154
+ "<|tt|>": 50351,
155
+ "<|uk|>": 50280,
156
+ "<|ur|>": 50290,
157
+ "<|uz|>": 50337,
158
+ "<|vi|>": 50278,
159
+ "<|yi|>": 50335,
160
+ "<|yo|>": 50325,
161
+ "<|zh|>": 50260
162
+ },
163
+ "language": "ar",
164
+ "max_initial_timestamp_index": 50,
165
+ "max_length": 448,
166
+ "no_timestamps_token_id": 50363,
167
+ "pad_token_id": 50257,
168
+ "prev_sot_token_id": 50361,
169
+ "return_timestamps": false,
170
+ "suppress_tokens": [
171
+ 1,
172
+ 2,
173
+ 7,
174
+ 8,
175
+ 9,
176
+ 10,
177
+ 14,
178
+ 25,
179
+ 26,
180
+ 27,
181
+ 28,
182
+ 29,
183
+ 31,
184
+ 58,
185
+ 59,
186
+ 60,
187
+ 61,
188
+ 62,
189
+ 63,
190
+ 90,
191
+ 91,
192
+ 92,
193
+ 93,
194
+ 359,
195
+ 503,
196
+ 522,
197
+ 542,
198
+ 873,
199
+ 893,
200
+ 902,
201
+ 918,
202
+ 922,
203
+ 931,
204
+ 1350,
205
+ 1853,
206
+ 1982,
207
+ 2460,
208
+ 2627,
209
+ 3246,
210
+ 3253,
211
+ 3268,
212
+ 3536,
213
+ 3846,
214
+ 3961,
215
+ 4183,
216
+ 4667,
217
+ 6585,
218
+ 6647,
219
+ 7273,
220
+ 9061,
221
+ 9383,
222
+ 10428,
223
+ 10929,
224
+ 11938,
225
+ 12033,
226
+ 12331,
227
+ 12562,
228
+ 13793,
229
+ 14157,
230
+ 14635,
231
+ 15265,
232
+ 15618,
233
+ 16553,
234
+ 16604,
235
+ 18362,
236
+ 18956,
237
+ 20075,
238
+ 21675,
239
+ 22520,
240
+ 26130,
241
+ 26161,
242
+ 26435,
243
+ 28279,
244
+ 29464,
245
+ 31650,
246
+ 32302,
247
+ 32470,
248
+ 36865,
249
+ 42863,
250
+ 47425,
251
+ 49870,
252
+ 50254,
253
+ 50258,
254
+ 50358,
255
+ 50359,
256
+ 50360,
257
+ 50361,
258
+ 50362
259
+ ],
260
+ "task_to_id": {
261
+ "transcribe": 50359,
262
+ "translate": 50358
263
+ },
264
+ "transformers_version": "4.40.1"
265
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:93f4e85472794bbcbcfc4dec49c7df91327dc88187fe9bd08402d262a85f4793
3
+ size 966995080
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:183a575b1c7bae52a385aaf65d40648fefbe2294ff323ee050506238661dfb4f
3
+ size 1925064044
preprocessor_config.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "chunk_length": 30,
3
+ "feature_extractor_type": "WhisperFeatureExtractor",
4
+ "feature_size": 80,
5
+ "hop_length": 160,
6
+ "n_fft": 400,
7
+ "n_samples": 480000,
8
+ "nb_max_frames": 3000,
9
+ "padding_side": "right",
10
+ "padding_value": 0.0,
11
+ "processor_class": "WhisperProcessor",
12
+ "return_attention_mask": false,
13
+ "sampling_rate": 16000
14
+ }
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d19a900bd07745422be13db39f2e6bf8f522c128c426fca823a48a96353192a
3
+ size 14244
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4388719db15f0ccd2f86a22b779cd689681f42021cea574bf093635775ad17df
3
+ size 1064
trainer_state.json ADDED
@@ -0,0 +1,821 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 81.30954374833746,
3
+ "best_model_checkpoint": "/scratch/p310333/whisper-small-dialect_all_seed84/checkpoint-1750",
4
+ "epoch": 0.3017137340091721,
5
+ "eval_steps": 250,
6
+ "global_step": 2500,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.003017137340091721,
13
+ "grad_norm": 65.5420150756836,
14
+ "learning_rate": 5.000000000000001e-07,
15
+ "loss": 4.9187,
16
+ "step": 25
17
+ },
18
+ {
19
+ "epoch": 0.006034274680183442,
20
+ "grad_norm": 30.61561393737793,
21
+ "learning_rate": 1.0000000000000002e-06,
22
+ "loss": 4.0023,
23
+ "step": 50
24
+ },
25
+ {
26
+ "epoch": 0.009051412020275163,
27
+ "grad_norm": 29.28253173828125,
28
+ "learning_rate": 1.5e-06,
29
+ "loss": 3.4756,
30
+ "step": 75
31
+ },
32
+ {
33
+ "epoch": 0.012068549360366883,
34
+ "grad_norm": 28.802133560180664,
35
+ "learning_rate": 2.0000000000000003e-06,
36
+ "loss": 2.7255,
37
+ "step": 100
38
+ },
39
+ {
40
+ "epoch": 0.015085686700458605,
41
+ "grad_norm": 32.795570373535156,
42
+ "learning_rate": 2.5e-06,
43
+ "loss": 2.4196,
44
+ "step": 125
45
+ },
46
+ {
47
+ "epoch": 0.018102824040550327,
48
+ "grad_norm": 23.08877182006836,
49
+ "learning_rate": 3e-06,
50
+ "loss": 2.2985,
51
+ "step": 150
52
+ },
53
+ {
54
+ "epoch": 0.021119961380642045,
55
+ "grad_norm": 26.61231231689453,
56
+ "learning_rate": 3.5e-06,
57
+ "loss": 2.3416,
58
+ "step": 175
59
+ },
60
+ {
61
+ "epoch": 0.024137098720733767,
62
+ "grad_norm": 19.47349739074707,
63
+ "learning_rate": 4.000000000000001e-06,
64
+ "loss": 2.1088,
65
+ "step": 200
66
+ },
67
+ {
68
+ "epoch": 0.02715423606082549,
69
+ "grad_norm": 33.220603942871094,
70
+ "learning_rate": 4.5e-06,
71
+ "loss": 2.0878,
72
+ "step": 225
73
+ },
74
+ {
75
+ "epoch": 0.03017137340091721,
76
+ "grad_norm": 27.01511001586914,
77
+ "learning_rate": 5e-06,
78
+ "loss": 2.1127,
79
+ "step": 250
80
+ },
81
+ {
82
+ "epoch": 0.03017137340091721,
83
+ "eval_cer": 75.04073104686333,
84
+ "eval_loss": 1.908042073249817,
85
+ "eval_runtime": 5452.7397,
86
+ "eval_samples_per_second": 3.039,
87
+ "eval_steps_per_second": 0.38,
88
+ "eval_wer": 89.40503049231042,
89
+ "step": 250
90
+ },
91
+ {
92
+ "epoch": 0.03318851074100893,
93
+ "grad_norm": 24.661699295043945,
94
+ "learning_rate": 5.500000000000001e-06,
95
+ "loss": 1.8497,
96
+ "step": 275
97
+ },
98
+ {
99
+ "epoch": 0.036205648081100654,
100
+ "grad_norm": 24.292240142822266,
101
+ "learning_rate": 6e-06,
102
+ "loss": 1.8874,
103
+ "step": 300
104
+ },
105
+ {
106
+ "epoch": 0.039222785421192376,
107
+ "grad_norm": 28.39283561706543,
108
+ "learning_rate": 6.5000000000000004e-06,
109
+ "loss": 1.7338,
110
+ "step": 325
111
+ },
112
+ {
113
+ "epoch": 0.04223992276128409,
114
+ "grad_norm": 28.00855827331543,
115
+ "learning_rate": 7e-06,
116
+ "loss": 1.6565,
117
+ "step": 350
118
+ },
119
+ {
120
+ "epoch": 0.04525706010137581,
121
+ "grad_norm": 29.236988067626953,
122
+ "learning_rate": 7.500000000000001e-06,
123
+ "loss": 1.471,
124
+ "step": 375
125
+ },
126
+ {
127
+ "epoch": 0.048274197441467534,
128
+ "grad_norm": 21.41299057006836,
129
+ "learning_rate": 8.000000000000001e-06,
130
+ "loss": 1.377,
131
+ "step": 400
132
+ },
133
+ {
134
+ "epoch": 0.051291334781559256,
135
+ "grad_norm": 25.506296157836914,
136
+ "learning_rate": 8.5e-06,
137
+ "loss": 1.3921,
138
+ "step": 425
139
+ },
140
+ {
141
+ "epoch": 0.05430847212165098,
142
+ "grad_norm": 29.994136810302734,
143
+ "learning_rate": 9e-06,
144
+ "loss": 1.4203,
145
+ "step": 450
146
+ },
147
+ {
148
+ "epoch": 0.0573256094617427,
149
+ "grad_norm": 30.166982650756836,
150
+ "learning_rate": 9.5e-06,
151
+ "loss": 1.4172,
152
+ "step": 475
153
+ },
154
+ {
155
+ "epoch": 0.06034274680183442,
156
+ "grad_norm": 26.911828994750977,
157
+ "learning_rate": 1e-05,
158
+ "loss": 1.409,
159
+ "step": 500
160
+ },
161
+ {
162
+ "epoch": 0.06034274680183442,
163
+ "eval_cer": 63.28322408864283,
164
+ "eval_loss": 1.46658456325531,
165
+ "eval_runtime": 3790.6529,
166
+ "eval_samples_per_second": 4.372,
167
+ "eval_steps_per_second": 0.547,
168
+ "eval_wer": 91.1192993172481,
169
+ "step": 500
170
+ },
171
+ {
172
+ "epoch": 0.06335988414192614,
173
+ "grad_norm": 29.040372848510742,
174
+ "learning_rate": 9.944444444444445e-06,
175
+ "loss": 1.5303,
176
+ "step": 525
177
+ },
178
+ {
179
+ "epoch": 0.06637702148201786,
180
+ "grad_norm": 24.52021598815918,
181
+ "learning_rate": 9.88888888888889e-06,
182
+ "loss": 1.3872,
183
+ "step": 550
184
+ },
185
+ {
186
+ "epoch": 0.06939415882210959,
187
+ "grad_norm": 23.997676849365234,
188
+ "learning_rate": 9.833333333333333e-06,
189
+ "loss": 1.3784,
190
+ "step": 575
191
+ },
192
+ {
193
+ "epoch": 0.07241129616220131,
194
+ "grad_norm": 22.321500778198242,
195
+ "learning_rate": 9.777777777777779e-06,
196
+ "loss": 1.5472,
197
+ "step": 600
198
+ },
199
+ {
200
+ "epoch": 0.07542843350229303,
201
+ "grad_norm": 24.63616943359375,
202
+ "learning_rate": 9.722222222222223e-06,
203
+ "loss": 1.337,
204
+ "step": 625
205
+ },
206
+ {
207
+ "epoch": 0.07844557084238475,
208
+ "grad_norm": 26.10930824279785,
209
+ "learning_rate": 9.666666666666667e-06,
210
+ "loss": 1.4113,
211
+ "step": 650
212
+ },
213
+ {
214
+ "epoch": 0.08146270818247647,
215
+ "grad_norm": 24.377735137939453,
216
+ "learning_rate": 9.611111111111112e-06,
217
+ "loss": 1.4169,
218
+ "step": 675
219
+ },
220
+ {
221
+ "epoch": 0.08447984552256818,
222
+ "grad_norm": 28.207748413085938,
223
+ "learning_rate": 9.555555555555556e-06,
224
+ "loss": 1.4385,
225
+ "step": 700
226
+ },
227
+ {
228
+ "epoch": 0.0874969828626599,
229
+ "grad_norm": 22.47800636291504,
230
+ "learning_rate": 9.5e-06,
231
+ "loss": 1.4014,
232
+ "step": 725
233
+ },
234
+ {
235
+ "epoch": 0.09051412020275162,
236
+ "grad_norm": 23.965164184570312,
237
+ "learning_rate": 9.444444444444445e-06,
238
+ "loss": 1.4844,
239
+ "step": 750
240
+ },
241
+ {
242
+ "epoch": 0.09051412020275162,
243
+ "eval_cer": 61.70583887099572,
244
+ "eval_loss": 1.4198288917541504,
245
+ "eval_runtime": 3756.7708,
246
+ "eval_samples_per_second": 4.411,
247
+ "eval_steps_per_second": 0.552,
248
+ "eval_wer": 89.3991192205003,
249
+ "step": 750
250
+ },
251
+ {
252
+ "epoch": 0.09353125754284335,
253
+ "grad_norm": 21.481678009033203,
254
+ "learning_rate": 9.38888888888889e-06,
255
+ "loss": 1.3299,
256
+ "step": 775
257
+ },
258
+ {
259
+ "epoch": 0.09654839488293507,
260
+ "grad_norm": 18.25922393798828,
261
+ "learning_rate": 9.333333333333334e-06,
262
+ "loss": 1.382,
263
+ "step": 800
264
+ },
265
+ {
266
+ "epoch": 0.09956553222302679,
267
+ "grad_norm": 23.341800689697266,
268
+ "learning_rate": 9.277777777777778e-06,
269
+ "loss": 1.5862,
270
+ "step": 825
271
+ },
272
+ {
273
+ "epoch": 0.10258266956311851,
274
+ "grad_norm": 21.80385971069336,
275
+ "learning_rate": 9.222222222222224e-06,
276
+ "loss": 1.3633,
277
+ "step": 850
278
+ },
279
+ {
280
+ "epoch": 0.10559980690321023,
281
+ "grad_norm": 21.196138381958008,
282
+ "learning_rate": 9.166666666666666e-06,
283
+ "loss": 1.373,
284
+ "step": 875
285
+ },
286
+ {
287
+ "epoch": 0.10861694424330195,
288
+ "grad_norm": 26.68990135192871,
289
+ "learning_rate": 9.111111111111112e-06,
290
+ "loss": 1.4125,
291
+ "step": 900
292
+ },
293
+ {
294
+ "epoch": 0.11163408158339368,
295
+ "grad_norm": 25.68580436706543,
296
+ "learning_rate": 9.055555555555556e-06,
297
+ "loss": 1.3544,
298
+ "step": 925
299
+ },
300
+ {
301
+ "epoch": 0.1146512189234854,
302
+ "grad_norm": 21.313331604003906,
303
+ "learning_rate": 9e-06,
304
+ "loss": 1.3692,
305
+ "step": 950
306
+ },
307
+ {
308
+ "epoch": 0.11766835626357712,
309
+ "grad_norm": 26.997135162353516,
310
+ "learning_rate": 8.944444444444446e-06,
311
+ "loss": 1.4534,
312
+ "step": 975
313
+ },
314
+ {
315
+ "epoch": 0.12068549360366884,
316
+ "grad_norm": 26.953454971313477,
317
+ "learning_rate": 8.888888888888888e-06,
318
+ "loss": 1.4644,
319
+ "step": 1000
320
+ },
321
+ {
322
+ "epoch": 0.12068549360366884,
323
+ "eval_cer": 61.789563800659245,
324
+ "eval_loss": 1.3863773345947266,
325
+ "eval_runtime": 3890.4048,
326
+ "eval_samples_per_second": 4.26,
327
+ "eval_steps_per_second": 0.533,
328
+ "eval_wer": 90.75083004108335,
329
+ "step": 1000
330
+ },
331
+ {
332
+ "epoch": 0.12370263094376056,
333
+ "grad_norm": 24.120258331298828,
334
+ "learning_rate": 8.833333333333334e-06,
335
+ "loss": 1.4748,
336
+ "step": 1025
337
+ },
338
+ {
339
+ "epoch": 0.12671976828385229,
340
+ "grad_norm": 26.698230743408203,
341
+ "learning_rate": 8.777777777777778e-06,
342
+ "loss": 1.4389,
343
+ "step": 1050
344
+ },
345
+ {
346
+ "epoch": 0.129736905623944,
347
+ "grad_norm": 31.574880599975586,
348
+ "learning_rate": 8.722222222222224e-06,
349
+ "loss": 1.3504,
350
+ "step": 1075
351
+ },
352
+ {
353
+ "epoch": 0.13275404296403573,
354
+ "grad_norm": 23.17409324645996,
355
+ "learning_rate": 8.666666666666668e-06,
356
+ "loss": 1.3387,
357
+ "step": 1100
358
+ },
359
+ {
360
+ "epoch": 0.13577118030412744,
361
+ "grad_norm": 27.488399505615234,
362
+ "learning_rate": 8.611111111111112e-06,
363
+ "loss": 1.4531,
364
+ "step": 1125
365
+ },
366
+ {
367
+ "epoch": 0.13878831764421917,
368
+ "grad_norm": 20.899181365966797,
369
+ "learning_rate": 8.555555555555556e-06,
370
+ "loss": 1.3062,
371
+ "step": 1150
372
+ },
373
+ {
374
+ "epoch": 0.14180545498431088,
375
+ "grad_norm": 26.456336975097656,
376
+ "learning_rate": 8.5e-06,
377
+ "loss": 1.399,
378
+ "step": 1175
379
+ },
380
+ {
381
+ "epoch": 0.14482259232440262,
382
+ "grad_norm": 18.58867645263672,
383
+ "learning_rate": 8.444444444444446e-06,
384
+ "loss": 1.2873,
385
+ "step": 1200
386
+ },
387
+ {
388
+ "epoch": 0.14783972966449432,
389
+ "grad_norm": 23.831193923950195,
390
+ "learning_rate": 8.38888888888889e-06,
391
+ "loss": 1.3038,
392
+ "step": 1225
393
+ },
394
+ {
395
+ "epoch": 0.15085686700458606,
396
+ "grad_norm": 27.28873634338379,
397
+ "learning_rate": 8.333333333333334e-06,
398
+ "loss": 1.386,
399
+ "step": 1250
400
+ },
401
+ {
402
+ "epoch": 0.15085686700458606,
403
+ "eval_cer": 66.82154521523341,
404
+ "eval_loss": 1.3677594661712646,
405
+ "eval_runtime": 4861.047,
406
+ "eval_samples_per_second": 3.409,
407
+ "eval_steps_per_second": 0.426,
408
+ "eval_wer": 93.2207564457493,
409
+ "step": 1250
410
+ },
411
+ {
412
+ "epoch": 0.15387400434467777,
413
+ "grad_norm": 24.41499137878418,
414
+ "learning_rate": 8.277777777777778e-06,
415
+ "loss": 1.4668,
416
+ "step": 1275
417
+ },
418
+ {
419
+ "epoch": 0.1568911416847695,
420
+ "grad_norm": 22.846540451049805,
421
+ "learning_rate": 8.222222222222222e-06,
422
+ "loss": 1.4219,
423
+ "step": 1300
424
+ },
425
+ {
426
+ "epoch": 0.1599082790248612,
427
+ "grad_norm": 23.818105697631836,
428
+ "learning_rate": 8.166666666666668e-06,
429
+ "loss": 1.2927,
430
+ "step": 1325
431
+ },
432
+ {
433
+ "epoch": 0.16292541636495295,
434
+ "grad_norm": 22.695980072021484,
435
+ "learning_rate": 8.111111111111112e-06,
436
+ "loss": 1.3731,
437
+ "step": 1350
438
+ },
439
+ {
440
+ "epoch": 0.16594255370504465,
441
+ "grad_norm": 26.142614364624023,
442
+ "learning_rate": 8.055555555555557e-06,
443
+ "loss": 1.3787,
444
+ "step": 1375
445
+ },
446
+ {
447
+ "epoch": 0.16895969104513636,
448
+ "grad_norm": 17.853261947631836,
449
+ "learning_rate": 8.000000000000001e-06,
450
+ "loss": 1.3649,
451
+ "step": 1400
452
+ },
453
+ {
454
+ "epoch": 0.1719768283852281,
455
+ "grad_norm": 23.05394172668457,
456
+ "learning_rate": 7.944444444444445e-06,
457
+ "loss": 1.3517,
458
+ "step": 1425
459
+ },
460
+ {
461
+ "epoch": 0.1749939657253198,
462
+ "grad_norm": 20.292495727539062,
463
+ "learning_rate": 7.88888888888889e-06,
464
+ "loss": 1.3752,
465
+ "step": 1450
466
+ },
467
+ {
468
+ "epoch": 0.17801110306541154,
469
+ "grad_norm": 19.236572265625,
470
+ "learning_rate": 7.833333333333333e-06,
471
+ "loss": 1.2467,
472
+ "step": 1475
473
+ },
474
+ {
475
+ "epoch": 0.18102824040550325,
476
+ "grad_norm": 23.87392807006836,
477
+ "learning_rate": 7.77777777777778e-06,
478
+ "loss": 1.3727,
479
+ "step": 1500
480
+ },
481
+ {
482
+ "epoch": 0.18102824040550325,
483
+ "eval_cer": 58.00949636814832,
484
+ "eval_loss": 1.350059151649475,
485
+ "eval_runtime": 3818.8115,
486
+ "eval_samples_per_second": 4.34,
487
+ "eval_steps_per_second": 0.543,
488
+ "eval_wer": 83.93020758416174,
489
+ "step": 1500
490
+ },
491
+ {
492
+ "epoch": 0.18404537774559498,
493
+ "grad_norm": 22.543930053710938,
494
+ "learning_rate": 7.722222222222223e-06,
495
+ "loss": 1.2917,
496
+ "step": 1525
497
+ },
498
+ {
499
+ "epoch": 0.1870625150856867,
500
+ "grad_norm": 22.561914443969727,
501
+ "learning_rate": 7.666666666666667e-06,
502
+ "loss": 1.2996,
503
+ "step": 1550
504
+ },
505
+ {
506
+ "epoch": 0.19007965242577843,
507
+ "grad_norm": 26.33420753479004,
508
+ "learning_rate": 7.611111111111111e-06,
509
+ "loss": 1.3633,
510
+ "step": 1575
511
+ },
512
+ {
513
+ "epoch": 0.19309678976587014,
514
+ "grad_norm": 20.257822036743164,
515
+ "learning_rate": 7.555555555555556e-06,
516
+ "loss": 1.2725,
517
+ "step": 1600
518
+ },
519
+ {
520
+ "epoch": 0.19611392710596187,
521
+ "grad_norm": 22.469804763793945,
522
+ "learning_rate": 7.500000000000001e-06,
523
+ "loss": 1.4868,
524
+ "step": 1625
525
+ },
526
+ {
527
+ "epoch": 0.19913106444605358,
528
+ "grad_norm": 23.334760665893555,
529
+ "learning_rate": 7.444444444444445e-06,
530
+ "loss": 1.2444,
531
+ "step": 1650
532
+ },
533
+ {
534
+ "epoch": 0.20214820178614532,
535
+ "grad_norm": 23.603315353393555,
536
+ "learning_rate": 7.38888888888889e-06,
537
+ "loss": 1.359,
538
+ "step": 1675
539
+ },
540
+ {
541
+ "epoch": 0.20516533912623702,
542
+ "grad_norm": 19.353214263916016,
543
+ "learning_rate": 7.333333333333333e-06,
544
+ "loss": 1.288,
545
+ "step": 1700
546
+ },
547
+ {
548
+ "epoch": 0.20818247646632876,
549
+ "grad_norm": 25.300758361816406,
550
+ "learning_rate": 7.277777777777778e-06,
551
+ "loss": 1.3908,
552
+ "step": 1725
553
+ },
554
+ {
555
+ "epoch": 0.21119961380642047,
556
+ "grad_norm": 21.219324111938477,
557
+ "learning_rate": 7.222222222222223e-06,
558
+ "loss": 1.3078,
559
+ "step": 1750
560
+ },
561
+ {
562
+ "epoch": 0.21119961380642047,
563
+ "eval_cer": 59.00212706578064,
564
+ "eval_loss": 1.3268232345581055,
565
+ "eval_runtime": 4148.3039,
566
+ "eval_samples_per_second": 3.995,
567
+ "eval_steps_per_second": 0.499,
568
+ "eval_wer": 81.30954374833746,
569
+ "step": 1750
570
+ },
571
+ {
572
+ "epoch": 0.2142167511465122,
573
+ "grad_norm": 23.480670928955078,
574
+ "learning_rate": 7.166666666666667e-06,
575
+ "loss": 1.32,
576
+ "step": 1775
577
+ },
578
+ {
579
+ "epoch": 0.2172338884866039,
580
+ "grad_norm": 25.72469139099121,
581
+ "learning_rate": 7.111111111111112e-06,
582
+ "loss": 1.2747,
583
+ "step": 1800
584
+ },
585
+ {
586
+ "epoch": 0.22025102582669562,
587
+ "grad_norm": 16.877225875854492,
588
+ "learning_rate": 7.055555555555557e-06,
589
+ "loss": 1.2142,
590
+ "step": 1825
591
+ },
592
+ {
593
+ "epoch": 0.22326816316678735,
594
+ "grad_norm": 22.20556640625,
595
+ "learning_rate": 7e-06,
596
+ "loss": 1.2684,
597
+ "step": 1850
598
+ },
599
+ {
600
+ "epoch": 0.22628530050687906,
601
+ "grad_norm": 15.339997291564941,
602
+ "learning_rate": 6.944444444444445e-06,
603
+ "loss": 1.2484,
604
+ "step": 1875
605
+ },
606
+ {
607
+ "epoch": 0.2293024378469708,
608
+ "grad_norm": 20.877666473388672,
609
+ "learning_rate": 6.88888888888889e-06,
610
+ "loss": 1.2099,
611
+ "step": 1900
612
+ },
613
+ {
614
+ "epoch": 0.2323195751870625,
615
+ "grad_norm": 28.011430740356445,
616
+ "learning_rate": 6.833333333333334e-06,
617
+ "loss": 1.2802,
618
+ "step": 1925
619
+ },
620
+ {
621
+ "epoch": 0.23533671252715424,
622
+ "grad_norm": 22.03005027770996,
623
+ "learning_rate": 6.777777777777779e-06,
624
+ "loss": 1.2924,
625
+ "step": 1950
626
+ },
627
+ {
628
+ "epoch": 0.23835384986724595,
629
+ "grad_norm": 22.320819854736328,
630
+ "learning_rate": 6.7222222222222235e-06,
631
+ "loss": 1.2741,
632
+ "step": 1975
633
+ },
634
+ {
635
+ "epoch": 0.24137098720733768,
636
+ "grad_norm": 20.871091842651367,
637
+ "learning_rate": 6.666666666666667e-06,
638
+ "loss": 1.2995,
639
+ "step": 2000
640
+ },
641
+ {
642
+ "epoch": 0.24137098720733768,
643
+ "eval_cer": 59.07340639779147,
644
+ "eval_loss": 1.3121392726898193,
645
+ "eval_runtime": 4204.9822,
646
+ "eval_samples_per_second": 3.941,
647
+ "eval_steps_per_second": 0.493,
648
+ "eval_wer": 86.45432064708723,
649
+ "step": 2000
650
+ },
651
+ {
652
+ "epoch": 0.2443881245474294,
653
+ "grad_norm": 23.991544723510742,
654
+ "learning_rate": 6.6111111111111115e-06,
655
+ "loss": 1.3492,
656
+ "step": 2025
657
+ },
658
+ {
659
+ "epoch": 0.24740526188752113,
660
+ "grad_norm": 22.863008499145508,
661
+ "learning_rate": 6.555555555555556e-06,
662
+ "loss": 1.308,
663
+ "step": 2050
664
+ },
665
+ {
666
+ "epoch": 0.25042239922761284,
667
+ "grad_norm": 17.017812728881836,
668
+ "learning_rate": 6.5000000000000004e-06,
669
+ "loss": 1.2589,
670
+ "step": 2075
671
+ },
672
+ {
673
+ "epoch": 0.25343953656770457,
674
+ "grad_norm": 18.99051284790039,
675
+ "learning_rate": 6.444444444444445e-06,
676
+ "loss": 1.2262,
677
+ "step": 2100
678
+ },
679
+ {
680
+ "epoch": 0.2564566739077963,
681
+ "grad_norm": 24.77310562133789,
682
+ "learning_rate": 6.3888888888888885e-06,
683
+ "loss": 1.2641,
684
+ "step": 2125
685
+ },
686
+ {
687
+ "epoch": 0.259473811247888,
688
+ "grad_norm": 25.287763595581055,
689
+ "learning_rate": 6.333333333333333e-06,
690
+ "loss": 1.2945,
691
+ "step": 2150
692
+ },
693
+ {
694
+ "epoch": 0.2624909485879797,
695
+ "grad_norm": 25.328964233398438,
696
+ "learning_rate": 6.277777777777778e-06,
697
+ "loss": 1.4174,
698
+ "step": 2175
699
+ },
700
+ {
701
+ "epoch": 0.26550808592807146,
702
+ "grad_norm": 21.193706512451172,
703
+ "learning_rate": 6.222222222222223e-06,
704
+ "loss": 1.3045,
705
+ "step": 2200
706
+ },
707
+ {
708
+ "epoch": 0.2685252232681632,
709
+ "grad_norm": 25.26439094543457,
710
+ "learning_rate": 6.166666666666667e-06,
711
+ "loss": 1.2792,
712
+ "step": 2225
713
+ },
714
+ {
715
+ "epoch": 0.2715423606082549,
716
+ "grad_norm": 18.658802032470703,
717
+ "learning_rate": 6.111111111111112e-06,
718
+ "loss": 1.3895,
719
+ "step": 2250
720
+ },
721
+ {
722
+ "epoch": 0.2715423606082549,
723
+ "eval_cer": 69.0187589099165,
724
+ "eval_loss": 1.296238660812378,
725
+ "eval_runtime": 4556.544,
726
+ "eval_samples_per_second": 3.637,
727
+ "eval_steps_per_second": 0.455,
728
+ "eval_wer": 94.42961153092088,
729
+ "step": 2250
730
+ },
731
+ {
732
+ "epoch": 0.2745594979483466,
733
+ "grad_norm": 19.182533264160156,
734
+ "learning_rate": 6.055555555555555e-06,
735
+ "loss": 1.3048,
736
+ "step": 2275
737
+ },
738
+ {
739
+ "epoch": 0.27757663528843834,
740
+ "grad_norm": 21.095041275024414,
741
+ "learning_rate": 6e-06,
742
+ "loss": 1.2783,
743
+ "step": 2300
744
+ },
745
+ {
746
+ "epoch": 0.28059377262853,
747
+ "grad_norm": 18.291179656982422,
748
+ "learning_rate": 5.944444444444445e-06,
749
+ "loss": 1.2607,
750
+ "step": 2325
751
+ },
752
+ {
753
+ "epoch": 0.28361090996862176,
754
+ "grad_norm": 16.70252227783203,
755
+ "learning_rate": 5.88888888888889e-06,
756
+ "loss": 1.2578,
757
+ "step": 2350
758
+ },
759
+ {
760
+ "epoch": 0.2866280473087135,
761
+ "grad_norm": 27.941680908203125,
762
+ "learning_rate": 5.833333333333334e-06,
763
+ "loss": 1.3026,
764
+ "step": 2375
765
+ },
766
+ {
767
+ "epoch": 0.28964518464880523,
768
+ "grad_norm": 26.314424514770508,
769
+ "learning_rate": 5.777777777777778e-06,
770
+ "loss": 1.2207,
771
+ "step": 2400
772
+ },
773
+ {
774
+ "epoch": 0.2926623219888969,
775
+ "grad_norm": 23.71686553955078,
776
+ "learning_rate": 5.722222222222222e-06,
777
+ "loss": 1.2565,
778
+ "step": 2425
779
+ },
780
+ {
781
+ "epoch": 0.29567945932898865,
782
+ "grad_norm": 24.66400909423828,
783
+ "learning_rate": 5.666666666666667e-06,
784
+ "loss": 1.3363,
785
+ "step": 2450
786
+ },
787
+ {
788
+ "epoch": 0.2986965966690804,
789
+ "grad_norm": 24.745983123779297,
790
+ "learning_rate": 5.611111111111112e-06,
791
+ "loss": 1.2785,
792
+ "step": 2475
793
+ },
794
+ {
795
+ "epoch": 0.3017137340091721,
796
+ "grad_norm": 23.765567779541016,
797
+ "learning_rate": 5.555555555555557e-06,
798
+ "loss": 1.3658,
799
+ "step": 2500
800
+ },
801
+ {
802
+ "epoch": 0.3017137340091721,
803
+ "eval_cer": 60.034545961969265,
804
+ "eval_loss": 1.2828303575515747,
805
+ "eval_runtime": 4164.8499,
806
+ "eval_samples_per_second": 3.979,
807
+ "eval_steps_per_second": 0.497,
808
+ "eval_wer": 84.99325129801677,
809
+ "step": 2500
810
+ }
811
+ ],
812
+ "logging_steps": 25,
813
+ "max_steps": 5000,
814
+ "num_input_tokens_seen": 0,
815
+ "num_train_epochs": 1,
816
+ "save_steps": 250,
817
+ "total_flos": 5.7717080064e+18,
818
+ "train_batch_size": 8,
819
+ "trial_name": null,
820
+ "trial_params": null
821
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a4d80345da3e67ae6f5163b9180309d128ac7168dccb87fb5e3fadd8ad567e3
3
+ size 5240