rosenkrumov95 commited on
Commit
a9e865c
·
verified ·
1 Parent(s): 7c4b14e

Upload 9 files

Browse files
config.json ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": null,
11
+ "bos_token_id": 50257,
12
+ "classifier_proj_size": 256,
13
+ "d_model": 768,
14
+ "decoder_attention_heads": 12,
15
+ "decoder_ffn_dim": 3072,
16
+ "decoder_layerdrop": 0.0,
17
+ "decoder_layers": 12,
18
+ "decoder_start_token_id": 50258,
19
+ "dropout": 0.0,
20
+ "encoder_attention_heads": 12,
21
+ "encoder_ffn_dim": 3072,
22
+ "encoder_layerdrop": 0.0,
23
+ "encoder_layers": 12,
24
+ "eos_token_id": 50257,
25
+ "forced_decoder_ids": [
26
+ [
27
+ 1,
28
+ 50259
29
+ ],
30
+ [
31
+ 2,
32
+ 50359
33
+ ],
34
+ [
35
+ 3,
36
+ 50363
37
+ ]
38
+ ],
39
+ "init_std": 0.02,
40
+ "is_encoder_decoder": true,
41
+ "mask_feature_length": 10,
42
+ "mask_feature_min_masks": 0,
43
+ "mask_feature_prob": 0.0,
44
+ "mask_time_length": 10,
45
+ "mask_time_min_masks": 2,
46
+ "mask_time_prob": 0.05,
47
+ "max_length": null,
48
+ "max_source_positions": 1500,
49
+ "max_target_positions": 448,
50
+ "median_filter_width": 7,
51
+ "model_type": "whisper",
52
+ "num_hidden_layers": 12,
53
+ "num_mel_bins": 80,
54
+ "pad_token_id": 50257,
55
+ "scale_embedding": false,
56
+ "torch_dtype": "float32",
57
+ "transformers_version": "4.47.0",
58
+ "use_cache": true,
59
+ "use_weighted_layer_sum": false,
60
+ "vocab_size": 51865
61
+ }
generation_config.json ADDED
@@ -0,0 +1,254 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "is_multilingual": true,
52
+ "lang_to_id": {
53
+ "<|af|>": 50327,
54
+ "<|am|>": 50334,
55
+ "<|ar|>": 50272,
56
+ "<|as|>": 50350,
57
+ "<|az|>": 50304,
58
+ "<|ba|>": 50355,
59
+ "<|be|>": 50330,
60
+ "<|bg|>": 50292,
61
+ "<|bn|>": 50302,
62
+ "<|bo|>": 50347,
63
+ "<|br|>": 50309,
64
+ "<|bs|>": 50315,
65
+ "<|ca|>": 50270,
66
+ "<|cs|>": 50283,
67
+ "<|cy|>": 50297,
68
+ "<|da|>": 50285,
69
+ "<|de|>": 50261,
70
+ "<|el|>": 50281,
71
+ "<|en|>": 50259,
72
+ "<|es|>": 50262,
73
+ "<|et|>": 50307,
74
+ "<|eu|>": 50310,
75
+ "<|fa|>": 50300,
76
+ "<|fi|>": 50277,
77
+ "<|fo|>": 50338,
78
+ "<|fr|>": 50265,
79
+ "<|gl|>": 50319,
80
+ "<|gu|>": 50333,
81
+ "<|haw|>": 50352,
82
+ "<|ha|>": 50354,
83
+ "<|he|>": 50279,
84
+ "<|hi|>": 50276,
85
+ "<|hr|>": 50291,
86
+ "<|ht|>": 50339,
87
+ "<|hu|>": 50286,
88
+ "<|hy|>": 50312,
89
+ "<|id|>": 50275,
90
+ "<|is|>": 50311,
91
+ "<|it|>": 50274,
92
+ "<|ja|>": 50266,
93
+ "<|jw|>": 50356,
94
+ "<|ka|>": 50329,
95
+ "<|kk|>": 50316,
96
+ "<|km|>": 50323,
97
+ "<|kn|>": 50306,
98
+ "<|ko|>": 50264,
99
+ "<|la|>": 50294,
100
+ "<|lb|>": 50345,
101
+ "<|ln|>": 50353,
102
+ "<|lo|>": 50336,
103
+ "<|lt|>": 50293,
104
+ "<|lv|>": 50301,
105
+ "<|mg|>": 50349,
106
+ "<|mi|>": 50295,
107
+ "<|mk|>": 50308,
108
+ "<|ml|>": 50296,
109
+ "<|mn|>": 50314,
110
+ "<|mr|>": 50320,
111
+ "<|ms|>": 50282,
112
+ "<|mt|>": 50343,
113
+ "<|my|>": 50346,
114
+ "<|ne|>": 50313,
115
+ "<|nl|>": 50271,
116
+ "<|nn|>": 50342,
117
+ "<|no|>": 50288,
118
+ "<|oc|>": 50328,
119
+ "<|pa|>": 50321,
120
+ "<|pl|>": 50269,
121
+ "<|ps|>": 50340,
122
+ "<|pt|>": 50267,
123
+ "<|ro|>": 50284,
124
+ "<|ru|>": 50263,
125
+ "<|sa|>": 50344,
126
+ "<|sd|>": 50332,
127
+ "<|si|>": 50322,
128
+ "<|sk|>": 50298,
129
+ "<|sl|>": 50305,
130
+ "<|sn|>": 50324,
131
+ "<|so|>": 50326,
132
+ "<|sq|>": 50317,
133
+ "<|sr|>": 50303,
134
+ "<|su|>": 50357,
135
+ "<|sv|>": 50273,
136
+ "<|sw|>": 50318,
137
+ "<|ta|>": 50287,
138
+ "<|te|>": 50299,
139
+ "<|tg|>": 50331,
140
+ "<|th|>": 50289,
141
+ "<|tk|>": 50341,
142
+ "<|tl|>": 50348,
143
+ "<|tr|>": 50268,
144
+ "<|tt|>": 50351,
145
+ "<|uk|>": 50280,
146
+ "<|ur|>": 50290,
147
+ "<|uz|>": 50337,
148
+ "<|vi|>": 50278,
149
+ "<|yi|>": 50335,
150
+ "<|yo|>": 50325,
151
+ "<|zh|>": 50260
152
+ },
153
+ "language": "bulgarian",
154
+ "max_initial_timestamp_index": 50,
155
+ "max_length": 448,
156
+ "no_timestamps_token_id": 50363,
157
+ "pad_token_id": 50257,
158
+ "prev_sot_token_id": 50361,
159
+ "return_timestamps": false,
160
+ "suppress_tokens": [
161
+ 1,
162
+ 2,
163
+ 7,
164
+ 8,
165
+ 9,
166
+ 10,
167
+ 14,
168
+ 25,
169
+ 26,
170
+ 27,
171
+ 28,
172
+ 29,
173
+ 31,
174
+ 58,
175
+ 59,
176
+ 60,
177
+ 61,
178
+ 62,
179
+ 63,
180
+ 90,
181
+ 91,
182
+ 92,
183
+ 93,
184
+ 359,
185
+ 503,
186
+ 522,
187
+ 542,
188
+ 873,
189
+ 893,
190
+ 902,
191
+ 918,
192
+ 922,
193
+ 931,
194
+ 1350,
195
+ 1853,
196
+ 1982,
197
+ 2460,
198
+ 2627,
199
+ 3246,
200
+ 3253,
201
+ 3268,
202
+ 3536,
203
+ 3846,
204
+ 3961,
205
+ 4183,
206
+ 4667,
207
+ 6585,
208
+ 6647,
209
+ 7273,
210
+ 9061,
211
+ 9383,
212
+ 10428,
213
+ 10929,
214
+ 11938,
215
+ 12033,
216
+ 12331,
217
+ 12562,
218
+ 13793,
219
+ 14157,
220
+ 14635,
221
+ 15265,
222
+ 15618,
223
+ 16553,
224
+ 16604,
225
+ 18362,
226
+ 18956,
227
+ 20075,
228
+ 21675,
229
+ 22520,
230
+ 26130,
231
+ 26161,
232
+ 26435,
233
+ 28279,
234
+ 29464,
235
+ 31650,
236
+ 32302,
237
+ 32470,
238
+ 36865,
239
+ 42863,
240
+ 47425,
241
+ 49870,
242
+ 50254,
243
+ 50258,
244
+ 50360,
245
+ 50361,
246
+ 50362
247
+ ],
248
+ "task": "transcribe",
249
+ "task_to_id": {
250
+ "transcribe": 50359,
251
+ "translate": 50358
252
+ },
253
+ "transformers_version": "4.47.0"
254
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:90b99036b8e92e9d6e9eacbfb0dd38595ac9510245403feed2b7dc738e7748c1
3
+ size 966995080
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84f6727cf5e9974086f2913883488fcc1472a5f1a2a6396d7a06fa9eeb65d649
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 sha256:27cdf2edd39f57a70573d9ff0027b58248741fcc4a77b968063bd6a9c61fd866
3
+ size 14244
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:60a85b8792c229b674da5f03812474e0e5d22e2f53b15c91ca077ac96c50df87
3
+ size 1064
trainer_state.json ADDED
@@ -0,0 +1,900 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 37.15599055412812,
3
+ "best_model_checkpoint": "./whisper-small-finetuned/checkpoint-6000",
4
+ "epoch": 1.20825,
5
+ "eval_steps": 2000,
6
+ "global_step": 6000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.00625,
13
+ "grad_norm": 11.4181547164917,
14
+ "learning_rate": 4.7000000000000005e-07,
15
+ "loss": 1.6163,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 0.0125,
20
+ "grad_norm": 7.794241428375244,
21
+ "learning_rate": 9.7e-07,
22
+ "loss": 1.3889,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 0.01875,
27
+ "grad_norm": 8.827476501464844,
28
+ "learning_rate": 1.4700000000000001e-06,
29
+ "loss": 1.2801,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 0.025,
34
+ "grad_norm": 6.476517200469971,
35
+ "learning_rate": 1.97e-06,
36
+ "loss": 1.2292,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 0.03125,
41
+ "grad_norm": 7.033958435058594,
42
+ "learning_rate": 2.47e-06,
43
+ "loss": 1.0831,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 0.0375,
48
+ "grad_norm": 6.324102401733398,
49
+ "learning_rate": 2.97e-06,
50
+ "loss": 1.0518,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 0.04375,
55
+ "grad_norm": 6.920722961425781,
56
+ "learning_rate": 3.4700000000000002e-06,
57
+ "loss": 0.9768,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 0.05,
62
+ "grad_norm": 6.655612468719482,
63
+ "learning_rate": 3.97e-06,
64
+ "loss": 0.9377,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 0.05625,
69
+ "grad_norm": 8.39956283569336,
70
+ "learning_rate": 4.47e-06,
71
+ "loss": 0.9107,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 0.0625,
76
+ "grad_norm": 6.610105514526367,
77
+ "learning_rate": 4.970000000000001e-06,
78
+ "loss": 0.9365,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 0.06875,
83
+ "grad_norm": 7.250374794006348,
84
+ "learning_rate": 5.470000000000001e-06,
85
+ "loss": 0.8301,
86
+ "step": 550
87
+ },
88
+ {
89
+ "epoch": 0.075,
90
+ "grad_norm": 5.3105669021606445,
91
+ "learning_rate": 5.9700000000000004e-06,
92
+ "loss": 0.8703,
93
+ "step": 600
94
+ },
95
+ {
96
+ "epoch": 0.08125,
97
+ "grad_norm": 5.701041221618652,
98
+ "learning_rate": 6.470000000000001e-06,
99
+ "loss": 0.8235,
100
+ "step": 650
101
+ },
102
+ {
103
+ "epoch": 0.0875,
104
+ "grad_norm": 5.628653049468994,
105
+ "learning_rate": 6.97e-06,
106
+ "loss": 0.7949,
107
+ "step": 700
108
+ },
109
+ {
110
+ "epoch": 0.09375,
111
+ "grad_norm": 6.777932643890381,
112
+ "learning_rate": 7.4700000000000005e-06,
113
+ "loss": 0.7607,
114
+ "step": 750
115
+ },
116
+ {
117
+ "epoch": 0.1,
118
+ "grad_norm": 5.994093418121338,
119
+ "learning_rate": 7.970000000000002e-06,
120
+ "loss": 0.7144,
121
+ "step": 800
122
+ },
123
+ {
124
+ "epoch": 0.10625,
125
+ "grad_norm": 7.069298267364502,
126
+ "learning_rate": 8.47e-06,
127
+ "loss": 0.807,
128
+ "step": 850
129
+ },
130
+ {
131
+ "epoch": 0.1125,
132
+ "grad_norm": 4.62778902053833,
133
+ "learning_rate": 8.97e-06,
134
+ "loss": 0.7103,
135
+ "step": 900
136
+ },
137
+ {
138
+ "epoch": 0.11875,
139
+ "grad_norm": 5.539309501647949,
140
+ "learning_rate": 9.47e-06,
141
+ "loss": 0.7273,
142
+ "step": 950
143
+ },
144
+ {
145
+ "epoch": 0.125,
146
+ "grad_norm": 4.9156036376953125,
147
+ "learning_rate": 9.970000000000001e-06,
148
+ "loss": 0.6556,
149
+ "step": 1000
150
+ },
151
+ {
152
+ "epoch": 0.13125,
153
+ "grad_norm": 5.073310375213623,
154
+ "learning_rate": 9.932857142857145e-06,
155
+ "loss": 0.7311,
156
+ "step": 1050
157
+ },
158
+ {
159
+ "epoch": 0.1375,
160
+ "grad_norm": 6.0137553215026855,
161
+ "learning_rate": 9.861428571428572e-06,
162
+ "loss": 0.6972,
163
+ "step": 1100
164
+ },
165
+ {
166
+ "epoch": 0.14375,
167
+ "grad_norm": 6.463374137878418,
168
+ "learning_rate": 9.790000000000001e-06,
169
+ "loss": 0.6923,
170
+ "step": 1150
171
+ },
172
+ {
173
+ "epoch": 0.15,
174
+ "grad_norm": 4.6144185066223145,
175
+ "learning_rate": 9.71857142857143e-06,
176
+ "loss": 0.5852,
177
+ "step": 1200
178
+ },
179
+ {
180
+ "epoch": 0.15625,
181
+ "grad_norm": 6.625977516174316,
182
+ "learning_rate": 9.647142857142857e-06,
183
+ "loss": 0.6973,
184
+ "step": 1250
185
+ },
186
+ {
187
+ "epoch": 0.1625,
188
+ "grad_norm": 5.229981899261475,
189
+ "learning_rate": 9.575714285714286e-06,
190
+ "loss": 0.6833,
191
+ "step": 1300
192
+ },
193
+ {
194
+ "epoch": 0.16875,
195
+ "grad_norm": 4.967216491699219,
196
+ "learning_rate": 9.504285714285715e-06,
197
+ "loss": 0.7299,
198
+ "step": 1350
199
+ },
200
+ {
201
+ "epoch": 0.175,
202
+ "grad_norm": 5.4233551025390625,
203
+ "learning_rate": 9.432857142857143e-06,
204
+ "loss": 0.7372,
205
+ "step": 1400
206
+ },
207
+ {
208
+ "epoch": 0.18125,
209
+ "grad_norm": 5.350655555725098,
210
+ "learning_rate": 9.361428571428572e-06,
211
+ "loss": 0.6854,
212
+ "step": 1450
213
+ },
214
+ {
215
+ "epoch": 0.1875,
216
+ "grad_norm": 4.796384811401367,
217
+ "learning_rate": 9.29e-06,
218
+ "loss": 0.5997,
219
+ "step": 1500
220
+ },
221
+ {
222
+ "epoch": 0.19375,
223
+ "grad_norm": 4.787758827209473,
224
+ "learning_rate": 9.218571428571428e-06,
225
+ "loss": 0.6864,
226
+ "step": 1550
227
+ },
228
+ {
229
+ "epoch": 0.2,
230
+ "grad_norm": 4.423880100250244,
231
+ "learning_rate": 9.147142857142857e-06,
232
+ "loss": 0.6138,
233
+ "step": 1600
234
+ },
235
+ {
236
+ "epoch": 0.20625,
237
+ "grad_norm": 4.9922895431518555,
238
+ "learning_rate": 9.075714285714286e-06,
239
+ "loss": 0.6438,
240
+ "step": 1650
241
+ },
242
+ {
243
+ "epoch": 0.2125,
244
+ "grad_norm": 5.809079170227051,
245
+ "learning_rate": 9.004285714285715e-06,
246
+ "loss": 0.6138,
247
+ "step": 1700
248
+ },
249
+ {
250
+ "epoch": 0.21875,
251
+ "grad_norm": 4.862271785736084,
252
+ "learning_rate": 8.932857142857143e-06,
253
+ "loss": 0.6552,
254
+ "step": 1750
255
+ },
256
+ {
257
+ "epoch": 0.225,
258
+ "grad_norm": 4.329502105712891,
259
+ "learning_rate": 8.861428571428572e-06,
260
+ "loss": 0.6638,
261
+ "step": 1800
262
+ },
263
+ {
264
+ "epoch": 0.23125,
265
+ "grad_norm": 4.74673318862915,
266
+ "learning_rate": 8.79e-06,
267
+ "loss": 0.7205,
268
+ "step": 1850
269
+ },
270
+ {
271
+ "epoch": 0.2375,
272
+ "grad_norm": 5.6587114334106445,
273
+ "learning_rate": 8.71857142857143e-06,
274
+ "loss": 0.6283,
275
+ "step": 1900
276
+ },
277
+ {
278
+ "epoch": 0.24375,
279
+ "grad_norm": 4.025678634643555,
280
+ "learning_rate": 8.647142857142859e-06,
281
+ "loss": 0.5727,
282
+ "step": 1950
283
+ },
284
+ {
285
+ "epoch": 0.25,
286
+ "grad_norm": 5.050192356109619,
287
+ "learning_rate": 8.575714285714286e-06,
288
+ "loss": 0.67,
289
+ "step": 2000
290
+ },
291
+ {
292
+ "epoch": 0.25,
293
+ "eval_loss": 0.6608728766441345,
294
+ "eval_runtime": 3982.4181,
295
+ "eval_samples_per_second": 1.089,
296
+ "eval_steps_per_second": 0.136,
297
+ "eval_wer": 40.78878971053452,
298
+ "step": 2000
299
+ },
300
+ {
301
+ "epoch": 0.25625,
302
+ "grad_norm": 5.174773216247559,
303
+ "learning_rate": 8.504285714285715e-06,
304
+ "loss": 0.7296,
305
+ "step": 2050
306
+ },
307
+ {
308
+ "epoch": 0.2625,
309
+ "grad_norm": 4.204230308532715,
310
+ "learning_rate": 8.432857142857144e-06,
311
+ "loss": 0.6267,
312
+ "step": 2100
313
+ },
314
+ {
315
+ "epoch": 0.26875,
316
+ "grad_norm": 4.041508674621582,
317
+ "learning_rate": 8.361428571428573e-06,
318
+ "loss": 0.6318,
319
+ "step": 2150
320
+ },
321
+ {
322
+ "epoch": 0.275,
323
+ "grad_norm": 4.373353958129883,
324
+ "learning_rate": 8.29e-06,
325
+ "loss": 0.6095,
326
+ "step": 2200
327
+ },
328
+ {
329
+ "epoch": 0.28125,
330
+ "grad_norm": 5.110273361206055,
331
+ "learning_rate": 8.21857142857143e-06,
332
+ "loss": 0.6114,
333
+ "step": 2250
334
+ },
335
+ {
336
+ "epoch": 0.2875,
337
+ "grad_norm": 7.183779239654541,
338
+ "learning_rate": 8.147142857142858e-06,
339
+ "loss": 0.6142,
340
+ "step": 2300
341
+ },
342
+ {
343
+ "epoch": 0.29375,
344
+ "grad_norm": 4.0610551834106445,
345
+ "learning_rate": 8.075714285714287e-06,
346
+ "loss": 0.5791,
347
+ "step": 2350
348
+ },
349
+ {
350
+ "epoch": 0.3,
351
+ "grad_norm": 4.765223503112793,
352
+ "learning_rate": 8.004285714285715e-06,
353
+ "loss": 0.6245,
354
+ "step": 2400
355
+ },
356
+ {
357
+ "epoch": 0.30625,
358
+ "grad_norm": 5.459305763244629,
359
+ "learning_rate": 7.932857142857144e-06,
360
+ "loss": 0.6329,
361
+ "step": 2450
362
+ },
363
+ {
364
+ "epoch": 0.3125,
365
+ "grad_norm": 4.936705112457275,
366
+ "learning_rate": 7.861428571428573e-06,
367
+ "loss": 0.6021,
368
+ "step": 2500
369
+ },
370
+ {
371
+ "epoch": 0.31875,
372
+ "grad_norm": 4.05385684967041,
373
+ "learning_rate": 7.790000000000002e-06,
374
+ "loss": 0.6026,
375
+ "step": 2550
376
+ },
377
+ {
378
+ "epoch": 0.325,
379
+ "grad_norm": 4.764458179473877,
380
+ "learning_rate": 7.72e-06,
381
+ "loss": 0.6191,
382
+ "step": 2600
383
+ },
384
+ {
385
+ "epoch": 0.33125,
386
+ "grad_norm": 4.8262434005737305,
387
+ "learning_rate": 7.64857142857143e-06,
388
+ "loss": 0.6451,
389
+ "step": 2650
390
+ },
391
+ {
392
+ "epoch": 0.3375,
393
+ "grad_norm": 5.038938522338867,
394
+ "learning_rate": 7.577142857142857e-06,
395
+ "loss": 0.6515,
396
+ "step": 2700
397
+ },
398
+ {
399
+ "epoch": 0.34375,
400
+ "grad_norm": 6.03196907043457,
401
+ "learning_rate": 7.505714285714286e-06,
402
+ "loss": 0.6328,
403
+ "step": 2750
404
+ },
405
+ {
406
+ "epoch": 0.35,
407
+ "grad_norm": 4.219414710998535,
408
+ "learning_rate": 7.434285714285715e-06,
409
+ "loss": 0.6202,
410
+ "step": 2800
411
+ },
412
+ {
413
+ "epoch": 0.35625,
414
+ "grad_norm": 4.451348781585693,
415
+ "learning_rate": 7.362857142857144e-06,
416
+ "loss": 0.6193,
417
+ "step": 2850
418
+ },
419
+ {
420
+ "epoch": 0.3625,
421
+ "grad_norm": 4.656250953674316,
422
+ "learning_rate": 7.291428571428571e-06,
423
+ "loss": 0.5295,
424
+ "step": 2900
425
+ },
426
+ {
427
+ "epoch": 0.36875,
428
+ "grad_norm": 14.434591293334961,
429
+ "learning_rate": 7.22e-06,
430
+ "loss": 0.6179,
431
+ "step": 2950
432
+ },
433
+ {
434
+ "epoch": 0.375,
435
+ "grad_norm": 4.388996124267578,
436
+ "learning_rate": 7.148571428571429e-06,
437
+ "loss": 0.5695,
438
+ "step": 3000
439
+ },
440
+ {
441
+ "epoch": 0.38125,
442
+ "grad_norm": 4.346401214599609,
443
+ "learning_rate": 7.077142857142858e-06,
444
+ "loss": 0.5755,
445
+ "step": 3050
446
+ },
447
+ {
448
+ "epoch": 0.3875,
449
+ "grad_norm": 4.0902252197265625,
450
+ "learning_rate": 7.0057142857142865e-06,
451
+ "loss": 0.6107,
452
+ "step": 3100
453
+ },
454
+ {
455
+ "epoch": 0.39375,
456
+ "grad_norm": 4.089328289031982,
457
+ "learning_rate": 6.934285714285715e-06,
458
+ "loss": 0.5957,
459
+ "step": 3150
460
+ },
461
+ {
462
+ "epoch": 0.4,
463
+ "grad_norm": 5.215518951416016,
464
+ "learning_rate": 6.862857142857144e-06,
465
+ "loss": 0.5947,
466
+ "step": 3200
467
+ },
468
+ {
469
+ "epoch": 0.40625,
470
+ "grad_norm": 4.511919975280762,
471
+ "learning_rate": 6.791428571428572e-06,
472
+ "loss": 0.6286,
473
+ "step": 3250
474
+ },
475
+ {
476
+ "epoch": 0.4125,
477
+ "grad_norm": 4.459268569946289,
478
+ "learning_rate": 6.720000000000001e-06,
479
+ "loss": 0.6103,
480
+ "step": 3300
481
+ },
482
+ {
483
+ "epoch": 0.41875,
484
+ "grad_norm": 4.5081868171691895,
485
+ "learning_rate": 6.648571428571429e-06,
486
+ "loss": 0.6093,
487
+ "step": 3350
488
+ },
489
+ {
490
+ "epoch": 0.425,
491
+ "grad_norm": 5.564323425292969,
492
+ "learning_rate": 6.577142857142857e-06,
493
+ "loss": 0.5998,
494
+ "step": 3400
495
+ },
496
+ {
497
+ "epoch": 0.43125,
498
+ "grad_norm": 4.530568599700928,
499
+ "learning_rate": 6.505714285714286e-06,
500
+ "loss": 0.5917,
501
+ "step": 3450
502
+ },
503
+ {
504
+ "epoch": 0.4375,
505
+ "grad_norm": 5.0652079582214355,
506
+ "learning_rate": 6.434285714285715e-06,
507
+ "loss": 0.5956,
508
+ "step": 3500
509
+ },
510
+ {
511
+ "epoch": 0.44375,
512
+ "grad_norm": 5.318765163421631,
513
+ "learning_rate": 6.3628571428571426e-06,
514
+ "loss": 0.6545,
515
+ "step": 3550
516
+ },
517
+ {
518
+ "epoch": 0.45,
519
+ "grad_norm": 4.156691074371338,
520
+ "learning_rate": 6.2914285714285716e-06,
521
+ "loss": 0.6077,
522
+ "step": 3600
523
+ },
524
+ {
525
+ "epoch": 0.45625,
526
+ "grad_norm": 4.188990116119385,
527
+ "learning_rate": 6.220000000000001e-06,
528
+ "loss": 0.6151,
529
+ "step": 3650
530
+ },
531
+ {
532
+ "epoch": 0.4625,
533
+ "grad_norm": 4.593184471130371,
534
+ "learning_rate": 6.14857142857143e-06,
535
+ "loss": 0.6372,
536
+ "step": 3700
537
+ },
538
+ {
539
+ "epoch": 0.46875,
540
+ "grad_norm": 6.544657230377197,
541
+ "learning_rate": 6.077142857142858e-06,
542
+ "loss": 0.633,
543
+ "step": 3750
544
+ },
545
+ {
546
+ "epoch": 0.475,
547
+ "grad_norm": 4.2101521492004395,
548
+ "learning_rate": 6.005714285714286e-06,
549
+ "loss": 0.564,
550
+ "step": 3800
551
+ },
552
+ {
553
+ "epoch": 0.48125,
554
+ "grad_norm": 4.94673490524292,
555
+ "learning_rate": 5.934285714285715e-06,
556
+ "loss": 0.631,
557
+ "step": 3850
558
+ },
559
+ {
560
+ "epoch": 0.4875,
561
+ "grad_norm": 5.893890380859375,
562
+ "learning_rate": 5.862857142857143e-06,
563
+ "loss": 0.5977,
564
+ "step": 3900
565
+ },
566
+ {
567
+ "epoch": 0.49375,
568
+ "grad_norm": 4.555218696594238,
569
+ "learning_rate": 5.791428571428572e-06,
570
+ "loss": 0.6037,
571
+ "step": 3950
572
+ },
573
+ {
574
+ "epoch": 0.5,
575
+ "grad_norm": 4.622659683227539,
576
+ "learning_rate": 5.72e-06,
577
+ "loss": 0.6431,
578
+ "step": 4000
579
+ },
580
+ {
581
+ "epoch": 0.5,
582
+ "eval_loss": 0.5986924767494202,
583
+ "eval_runtime": 3981.5379,
584
+ "eval_samples_per_second": 1.09,
585
+ "eval_steps_per_second": 0.136,
586
+ "eval_wer": 40.029852519641764,
587
+ "step": 4000
588
+ },
589
+ {
590
+ "epoch": 0.50625,
591
+ "grad_norm": 5.098015785217285,
592
+ "learning_rate": 5.6485714285714285e-06,
593
+ "loss": 0.6914,
594
+ "step": 4050
595
+ },
596
+ {
597
+ "epoch": 0.5125,
598
+ "grad_norm": 5.3102312088012695,
599
+ "learning_rate": 5.5771428571428575e-06,
600
+ "loss": 0.6189,
601
+ "step": 4100
602
+ },
603
+ {
604
+ "epoch": 0.51875,
605
+ "grad_norm": 4.711716651916504,
606
+ "learning_rate": 5.5057142857142865e-06,
607
+ "loss": 0.6726,
608
+ "step": 4150
609
+ },
610
+ {
611
+ "epoch": 0.525,
612
+ "grad_norm": 4.459558963775635,
613
+ "learning_rate": 5.4342857142857155e-06,
614
+ "loss": 0.6484,
615
+ "step": 4200
616
+ },
617
+ {
618
+ "epoch": 0.53125,
619
+ "grad_norm": 5.451053619384766,
620
+ "learning_rate": 5.362857142857143e-06,
621
+ "loss": 0.5951,
622
+ "step": 4250
623
+ },
624
+ {
625
+ "epoch": 0.5375,
626
+ "grad_norm": 3.955195665359497,
627
+ "learning_rate": 5.291428571428572e-06,
628
+ "loss": 0.5998,
629
+ "step": 4300
630
+ },
631
+ {
632
+ "epoch": 1.002,
633
+ "grad_norm": 4.236627578735352,
634
+ "learning_rate": 5.220000000000001e-06,
635
+ "loss": 0.572,
636
+ "step": 4350
637
+ },
638
+ {
639
+ "epoch": 1.00825,
640
+ "grad_norm": 5.792555332183838,
641
+ "learning_rate": 5.14857142857143e-06,
642
+ "loss": 0.5485,
643
+ "step": 4400
644
+ },
645
+ {
646
+ "epoch": 1.0145,
647
+ "grad_norm": 4.863269329071045,
648
+ "learning_rate": 5.077142857142857e-06,
649
+ "loss": 0.538,
650
+ "step": 4450
651
+ },
652
+ {
653
+ "epoch": 1.02075,
654
+ "grad_norm": 5.243031024932861,
655
+ "learning_rate": 5.005714285714286e-06,
656
+ "loss": 0.6561,
657
+ "step": 4500
658
+ },
659
+ {
660
+ "epoch": 1.027,
661
+ "grad_norm": 4.527797698974609,
662
+ "learning_rate": 4.934285714285715e-06,
663
+ "loss": 0.5889,
664
+ "step": 4550
665
+ },
666
+ {
667
+ "epoch": 1.03325,
668
+ "grad_norm": 4.516623020172119,
669
+ "learning_rate": 4.862857142857143e-06,
670
+ "loss": 0.5596,
671
+ "step": 4600
672
+ },
673
+ {
674
+ "epoch": 1.0395,
675
+ "grad_norm": 4.004039287567139,
676
+ "learning_rate": 4.7914285714285715e-06,
677
+ "loss": 0.5699,
678
+ "step": 4650
679
+ },
680
+ {
681
+ "epoch": 1.04575,
682
+ "grad_norm": 4.805927276611328,
683
+ "learning_rate": 4.7200000000000005e-06,
684
+ "loss": 0.5761,
685
+ "step": 4700
686
+ },
687
+ {
688
+ "epoch": 1.052,
689
+ "grad_norm": 4.1281962394714355,
690
+ "learning_rate": 4.648571428571429e-06,
691
+ "loss": 0.5827,
692
+ "step": 4750
693
+ },
694
+ {
695
+ "epoch": 1.05825,
696
+ "grad_norm": 4.829611778259277,
697
+ "learning_rate": 4.577142857142858e-06,
698
+ "loss": 0.5865,
699
+ "step": 4800
700
+ },
701
+ {
702
+ "epoch": 1.0645,
703
+ "grad_norm": 3.7667906284332275,
704
+ "learning_rate": 4.505714285714286e-06,
705
+ "loss": 0.5711,
706
+ "step": 4850
707
+ },
708
+ {
709
+ "epoch": 1.07075,
710
+ "grad_norm": 4.786715984344482,
711
+ "learning_rate": 4.434285714285715e-06,
712
+ "loss": 0.5728,
713
+ "step": 4900
714
+ },
715
+ {
716
+ "epoch": 1.077,
717
+ "grad_norm": 4.958069801330566,
718
+ "learning_rate": 4.362857142857143e-06,
719
+ "loss": 0.5611,
720
+ "step": 4950
721
+ },
722
+ {
723
+ "epoch": 1.08325,
724
+ "grad_norm": 3.8642232418060303,
725
+ "learning_rate": 4.291428571428572e-06,
726
+ "loss": 0.5695,
727
+ "step": 5000
728
+ },
729
+ {
730
+ "epoch": 1.0895,
731
+ "grad_norm": 4.423571586608887,
732
+ "learning_rate": 4.22e-06,
733
+ "loss": 0.5213,
734
+ "step": 5050
735
+ },
736
+ {
737
+ "epoch": 1.09575,
738
+ "grad_norm": 4.207879066467285,
739
+ "learning_rate": 4.148571428571429e-06,
740
+ "loss": 0.5132,
741
+ "step": 5100
742
+ },
743
+ {
744
+ "epoch": 1.102,
745
+ "grad_norm": 4.761569499969482,
746
+ "learning_rate": 4.0771428571428574e-06,
747
+ "loss": 0.5234,
748
+ "step": 5150
749
+ },
750
+ {
751
+ "epoch": 1.10825,
752
+ "grad_norm": 4.476335525512695,
753
+ "learning_rate": 4.0057142857142864e-06,
754
+ "loss": 0.5797,
755
+ "step": 5200
756
+ },
757
+ {
758
+ "epoch": 1.1145,
759
+ "grad_norm": 3.8417840003967285,
760
+ "learning_rate": 3.934285714285715e-06,
761
+ "loss": 0.4683,
762
+ "step": 5250
763
+ },
764
+ {
765
+ "epoch": 1.12075,
766
+ "grad_norm": 3.265429735183716,
767
+ "learning_rate": 3.862857142857143e-06,
768
+ "loss": 0.4727,
769
+ "step": 5300
770
+ },
771
+ {
772
+ "epoch": 1.127,
773
+ "grad_norm": 4.57379674911499,
774
+ "learning_rate": 3.7914285714285722e-06,
775
+ "loss": 0.5032,
776
+ "step": 5350
777
+ },
778
+ {
779
+ "epoch": 1.13325,
780
+ "grad_norm": 5.103521823883057,
781
+ "learning_rate": 3.7200000000000004e-06,
782
+ "loss": 0.5116,
783
+ "step": 5400
784
+ },
785
+ {
786
+ "epoch": 1.1395,
787
+ "grad_norm": 4.07943058013916,
788
+ "learning_rate": 3.648571428571429e-06,
789
+ "loss": 0.4916,
790
+ "step": 5450
791
+ },
792
+ {
793
+ "epoch": 1.14575,
794
+ "grad_norm": 4.684885501861572,
795
+ "learning_rate": 3.5771428571428576e-06,
796
+ "loss": 0.4923,
797
+ "step": 5500
798
+ },
799
+ {
800
+ "epoch": 1.152,
801
+ "grad_norm": 3.8068909645080566,
802
+ "learning_rate": 3.505714285714286e-06,
803
+ "loss": 0.4273,
804
+ "step": 5550
805
+ },
806
+ {
807
+ "epoch": 1.15825,
808
+ "grad_norm": 4.139249324798584,
809
+ "learning_rate": 3.4342857142857143e-06,
810
+ "loss": 0.5505,
811
+ "step": 5600
812
+ },
813
+ {
814
+ "epoch": 1.1645,
815
+ "grad_norm": 5.06072998046875,
816
+ "learning_rate": 3.3628571428571433e-06,
817
+ "loss": 0.4919,
818
+ "step": 5650
819
+ },
820
+ {
821
+ "epoch": 1.17075,
822
+ "grad_norm": 4.353974342346191,
823
+ "learning_rate": 3.2914285714285715e-06,
824
+ "loss": 0.5165,
825
+ "step": 5700
826
+ },
827
+ {
828
+ "epoch": 1.177,
829
+ "grad_norm": 4.245412826538086,
830
+ "learning_rate": 3.2200000000000005e-06,
831
+ "loss": 0.543,
832
+ "step": 5750
833
+ },
834
+ {
835
+ "epoch": 1.18325,
836
+ "grad_norm": 3.9986581802368164,
837
+ "learning_rate": 3.1485714285714287e-06,
838
+ "loss": 0.5065,
839
+ "step": 5800
840
+ },
841
+ {
842
+ "epoch": 1.1895,
843
+ "grad_norm": 4.012422561645508,
844
+ "learning_rate": 3.0771428571428573e-06,
845
+ "loss": 0.4563,
846
+ "step": 5850
847
+ },
848
+ {
849
+ "epoch": 1.19575,
850
+ "grad_norm": 3.4883952140808105,
851
+ "learning_rate": 3.005714285714286e-06,
852
+ "loss": 0.517,
853
+ "step": 5900
854
+ },
855
+ {
856
+ "epoch": 1.202,
857
+ "grad_norm": 3.5395171642303467,
858
+ "learning_rate": 2.9342857142857144e-06,
859
+ "loss": 0.4525,
860
+ "step": 5950
861
+ },
862
+ {
863
+ "epoch": 1.20825,
864
+ "grad_norm": 4.436487674713135,
865
+ "learning_rate": 2.8628571428571435e-06,
866
+ "loss": 0.476,
867
+ "step": 6000
868
+ },
869
+ {
870
+ "epoch": 1.20825,
871
+ "eval_loss": 0.577191948890686,
872
+ "eval_runtime": 3956.9016,
873
+ "eval_samples_per_second": 1.096,
874
+ "eval_steps_per_second": 0.137,
875
+ "eval_wer": 37.15599055412812,
876
+ "step": 6000
877
+ }
878
+ ],
879
+ "logging_steps": 50,
880
+ "max_steps": 8000,
881
+ "num_input_tokens_seen": 0,
882
+ "num_train_epochs": 9223372036854775807,
883
+ "save_steps": 2000,
884
+ "stateful_callbacks": {
885
+ "TrainerControl": {
886
+ "args": {
887
+ "should_epoch_stop": false,
888
+ "should_evaluate": false,
889
+ "should_log": false,
890
+ "should_save": true,
891
+ "should_training_stop": false
892
+ },
893
+ "attributes": {}
894
+ }
895
+ },
896
+ "total_flos": 1.385036770295808e+19,
897
+ "train_batch_size": 8,
898
+ "trial_name": null,
899
+ "trial_params": null
900
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ca5b86c01b9fa879ab624f5b6f627e7ece683e215a902b36b5fcde88f953f08e
3
+ size 5496