StellaLJQ commited on
Commit
18db742
·
verified ·
1 Parent(s): 6acf6df

Upload folder using huggingface_hub

Browse files
whisper-GST-exp2/checkpoint-3000/config.json ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "openai/whisper-tiny",
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": 384,
14
+ "decoder_attention_heads": 6,
15
+ "decoder_ffn_dim": 1536,
16
+ "decoder_layerdrop": 0.0,
17
+ "decoder_layers": 4,
18
+ "decoder_start_token_id": 50258,
19
+ "dropout": 0.0,
20
+ "encoder_attention_heads": 6,
21
+ "encoder_ffn_dim": 1536,
22
+ "encoder_layerdrop": 0.0,
23
+ "encoder_layers": 4,
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": 4,
53
+ "num_mel_bins": 80,
54
+ "pad_token_id": 50257,
55
+ "scale_embedding": false,
56
+ "torch_dtype": "float32",
57
+ "transformers_version": "4.48.0",
58
+ "use_cache": true,
59
+ "use_weighted_layer_sum": false,
60
+ "vocab_size": 51873
61
+ }
whisper-GST-exp2/checkpoint-3000/generation_config.json ADDED
@@ -0,0 +1,248 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alignment_heads": [
3
+ [
4
+ 2,
5
+ 2
6
+ ],
7
+ [
8
+ 3,
9
+ 0
10
+ ],
11
+ [
12
+ 3,
13
+ 2
14
+ ],
15
+ [
16
+ 3,
17
+ 3
18
+ ],
19
+ [
20
+ 3,
21
+ 4
22
+ ],
23
+ [
24
+ 3,
25
+ 5
26
+ ]
27
+ ],
28
+ "begin_suppress_tokens": [
29
+ 220,
30
+ 50257
31
+ ],
32
+ "bos_token_id": 50257,
33
+ "decoder_start_token_id": 50258,
34
+ "eos_token_id": 50257,
35
+ "forced_decoder_ids": [
36
+ [
37
+ 1,
38
+ null
39
+ ],
40
+ [
41
+ 2,
42
+ 50359
43
+ ]
44
+ ],
45
+ "is_multilingual": true,
46
+ "lang_to_id": {
47
+ "<|af|>": 50327,
48
+ "<|am|>": 50334,
49
+ "<|ar|>": 50272,
50
+ "<|as|>": 50350,
51
+ "<|az|>": 50304,
52
+ "<|ba|>": 50355,
53
+ "<|be|>": 50330,
54
+ "<|bg|>": 50292,
55
+ "<|bn|>": 50302,
56
+ "<|bo|>": 50347,
57
+ "<|br|>": 50309,
58
+ "<|bs|>": 50315,
59
+ "<|ca|>": 50270,
60
+ "<|cs|>": 50283,
61
+ "<|cy|>": 50297,
62
+ "<|da|>": 50285,
63
+ "<|de|>": 50261,
64
+ "<|el|>": 50281,
65
+ "<|en|>": 50259,
66
+ "<|es|>": 50262,
67
+ "<|et|>": 50307,
68
+ "<|eu|>": 50310,
69
+ "<|fa|>": 50300,
70
+ "<|fi|>": 50277,
71
+ "<|fo|>": 50338,
72
+ "<|fr|>": 50265,
73
+ "<|gl|>": 50319,
74
+ "<|gu|>": 50333,
75
+ "<|haw|>": 50352,
76
+ "<|ha|>": 50354,
77
+ "<|he|>": 50279,
78
+ "<|hi|>": 50276,
79
+ "<|hr|>": 50291,
80
+ "<|ht|>": 50339,
81
+ "<|hu|>": 50286,
82
+ "<|hy|>": 50312,
83
+ "<|id|>": 50275,
84
+ "<|is|>": 50311,
85
+ "<|it|>": 50274,
86
+ "<|ja|>": 50266,
87
+ "<|jw|>": 50356,
88
+ "<|ka|>": 50329,
89
+ "<|kk|>": 50316,
90
+ "<|km|>": 50323,
91
+ "<|kn|>": 50306,
92
+ "<|ko|>": 50264,
93
+ "<|la|>": 50294,
94
+ "<|lb|>": 50345,
95
+ "<|ln|>": 50353,
96
+ "<|lo|>": 50336,
97
+ "<|lt|>": 50293,
98
+ "<|lv|>": 50301,
99
+ "<|mg|>": 50349,
100
+ "<|mi|>": 50295,
101
+ "<|mk|>": 50308,
102
+ "<|ml|>": 50296,
103
+ "<|mn|>": 50314,
104
+ "<|mr|>": 50320,
105
+ "<|ms|>": 50282,
106
+ "<|mt|>": 50343,
107
+ "<|my|>": 50346,
108
+ "<|ne|>": 50313,
109
+ "<|nl|>": 50271,
110
+ "<|nn|>": 50342,
111
+ "<|no|>": 50288,
112
+ "<|oc|>": 50328,
113
+ "<|pa|>": 50321,
114
+ "<|pl|>": 50269,
115
+ "<|ps|>": 50340,
116
+ "<|pt|>": 50267,
117
+ "<|ro|>": 50284,
118
+ "<|ru|>": 50263,
119
+ "<|sa|>": 50344,
120
+ "<|sd|>": 50332,
121
+ "<|si|>": 50322,
122
+ "<|sk|>": 50298,
123
+ "<|sl|>": 50305,
124
+ "<|sn|>": 50324,
125
+ "<|so|>": 50326,
126
+ "<|sq|>": 50317,
127
+ "<|sr|>": 50303,
128
+ "<|su|>": 50357,
129
+ "<|sv|>": 50273,
130
+ "<|sw|>": 50318,
131
+ "<|ta|>": 50287,
132
+ "<|te|>": 50299,
133
+ "<|tg|>": 50331,
134
+ "<|th|>": 50289,
135
+ "<|tk|>": 50341,
136
+ "<|tl|>": 50348,
137
+ "<|tr|>": 50268,
138
+ "<|tt|>": 50351,
139
+ "<|uk|>": 50280,
140
+ "<|ur|>": 50290,
141
+ "<|uz|>": 50337,
142
+ "<|vi|>": 50278,
143
+ "<|yi|>": 50335,
144
+ "<|yo|>": 50325,
145
+ "<|zh|>": 50260
146
+ },
147
+ "max_initial_timestamp_index": 50,
148
+ "max_length": 448,
149
+ "no_timestamps_token_id": 50363,
150
+ "pad_token_id": 50257,
151
+ "prev_sot_token_id": 50361,
152
+ "return_timestamps": false,
153
+ "suppress_tokens": [
154
+ 1,
155
+ 2,
156
+ 7,
157
+ 8,
158
+ 9,
159
+ 10,
160
+ 14,
161
+ 25,
162
+ 26,
163
+ 27,
164
+ 28,
165
+ 29,
166
+ 31,
167
+ 58,
168
+ 59,
169
+ 60,
170
+ 61,
171
+ 62,
172
+ 63,
173
+ 90,
174
+ 91,
175
+ 92,
176
+ 93,
177
+ 359,
178
+ 503,
179
+ 522,
180
+ 542,
181
+ 873,
182
+ 893,
183
+ 902,
184
+ 918,
185
+ 922,
186
+ 931,
187
+ 1350,
188
+ 1853,
189
+ 1982,
190
+ 2460,
191
+ 2627,
192
+ 3246,
193
+ 3253,
194
+ 3268,
195
+ 3536,
196
+ 3846,
197
+ 3961,
198
+ 4183,
199
+ 4667,
200
+ 6585,
201
+ 6647,
202
+ 7273,
203
+ 9061,
204
+ 9383,
205
+ 10428,
206
+ 10929,
207
+ 11938,
208
+ 12033,
209
+ 12331,
210
+ 12562,
211
+ 13793,
212
+ 14157,
213
+ 14635,
214
+ 15265,
215
+ 15618,
216
+ 16553,
217
+ 16604,
218
+ 18362,
219
+ 18956,
220
+ 20075,
221
+ 21675,
222
+ 22520,
223
+ 26130,
224
+ 26161,
225
+ 26435,
226
+ 28279,
227
+ 29464,
228
+ 31650,
229
+ 32302,
230
+ 32470,
231
+ 36865,
232
+ 42863,
233
+ 47425,
234
+ 49870,
235
+ 50254,
236
+ 50258,
237
+ 50358,
238
+ 50359,
239
+ 50360,
240
+ 50361,
241
+ 50362
242
+ ],
243
+ "task_to_id": {
244
+ "transcribe": 50359,
245
+ "translate": 50358
246
+ },
247
+ "transformers_version": "4.48.0"
248
+ }
whisper-GST-exp2/checkpoint-3000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5a5b701cc4c13756810cbfff8216e2ae2f87cbc7ee377724919d31e54eab0fb6
3
+ size 160466696
whisper-GST-exp2/checkpoint-3000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c7f66fa70f1b667f508813143f873d8cfe45f7c3e75b4e33051ab34148ca0564
3
+ size 321073018
whisper-GST-exp2/checkpoint-3000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7a5642a9f71a477b0740aaf1fafc639b58e471dff191c5a467c151f2a094d84c
3
+ size 14244
whisper-GST-exp2/checkpoint-3000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d069d021e3ca87c952c1cbefefa02430980ec1efa75a58a8bd98140e99e3bc1
3
+ size 1064
whisper-GST-exp2/checkpoint-3000/trainer_state.json ADDED
@@ -0,0 +1,513 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.53125,
3
+ "best_model_checkpoint": "./whisper-GST-exp2/checkpoint-3000",
4
+ "epoch": 41.666666666666664,
5
+ "eval_steps": 500,
6
+ "global_step": 3000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.6944444444444444,
13
+ "grad_norm": 5.084019660949707,
14
+ "learning_rate": 9.800000000000001e-07,
15
+ "loss": 0.0377,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 1.3888888888888888,
20
+ "grad_norm": 4.491146564483643,
21
+ "learning_rate": 1.98e-06,
22
+ "loss": 0.0402,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 2.0833333333333335,
27
+ "grad_norm": 6.822654724121094,
28
+ "learning_rate": 2.9800000000000003e-06,
29
+ "loss": 0.0411,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 2.7777777777777777,
34
+ "grad_norm": 2.353851318359375,
35
+ "learning_rate": 3.980000000000001e-06,
36
+ "loss": 0.0429,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 3.4722222222222223,
41
+ "grad_norm": 3.9458818435668945,
42
+ "learning_rate": 4.980000000000001e-06,
43
+ "loss": 0.0425,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 4.166666666666667,
48
+ "grad_norm": 6.142700672149658,
49
+ "learning_rate": 5.98e-06,
50
+ "loss": 0.045,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 4.861111111111111,
55
+ "grad_norm": 4.524245262145996,
56
+ "learning_rate": 6.98e-06,
57
+ "loss": 0.0455,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 5.555555555555555,
62
+ "grad_norm": 2.7861649990081787,
63
+ "learning_rate": 7.980000000000002e-06,
64
+ "loss": 0.0457,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 6.25,
69
+ "grad_norm": 5.412281513214111,
70
+ "learning_rate": 8.98e-06,
71
+ "loss": 0.045,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 6.944444444444445,
76
+ "grad_norm": 5.459417343139648,
77
+ "learning_rate": 9.960000000000001e-06,
78
+ "loss": 0.0502,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 6.944444444444445,
83
+ "eval_emotion_accuracy": 0.46875,
84
+ "eval_loss": 0.19364911317825317,
85
+ "eval_runtime": 22.6537,
86
+ "eval_samples_per_second": 12.713,
87
+ "eval_steps_per_second": 1.589,
88
+ "eval_wer": 8.969907407407407,
89
+ "step": 500
90
+ },
91
+ {
92
+ "epoch": 7.638888888888889,
93
+ "grad_norm": 3.5574755668640137,
94
+ "learning_rate": 9.893333333333334e-06,
95
+ "loss": 0.0469,
96
+ "step": 550
97
+ },
98
+ {
99
+ "epoch": 8.333333333333334,
100
+ "grad_norm": 4.717072010040283,
101
+ "learning_rate": 9.782222222222222e-06,
102
+ "loss": 0.0498,
103
+ "step": 600
104
+ },
105
+ {
106
+ "epoch": 9.027777777777779,
107
+ "grad_norm": 4.033562183380127,
108
+ "learning_rate": 9.671111111111112e-06,
109
+ "loss": 0.044,
110
+ "step": 650
111
+ },
112
+ {
113
+ "epoch": 9.722222222222221,
114
+ "grad_norm": 8.465327262878418,
115
+ "learning_rate": 9.56e-06,
116
+ "loss": 0.0434,
117
+ "step": 700
118
+ },
119
+ {
120
+ "epoch": 10.416666666666666,
121
+ "grad_norm": 3.5226399898529053,
122
+ "learning_rate": 9.44888888888889e-06,
123
+ "loss": 0.0358,
124
+ "step": 750
125
+ },
126
+ {
127
+ "epoch": 11.11111111111111,
128
+ "grad_norm": 1.9998486042022705,
129
+ "learning_rate": 9.33777777777778e-06,
130
+ "loss": 0.0421,
131
+ "step": 800
132
+ },
133
+ {
134
+ "epoch": 11.805555555555555,
135
+ "grad_norm": 2.3085238933563232,
136
+ "learning_rate": 9.226666666666668e-06,
137
+ "loss": 0.0412,
138
+ "step": 850
139
+ },
140
+ {
141
+ "epoch": 12.5,
142
+ "grad_norm": 2.7251176834106445,
143
+ "learning_rate": 9.115555555555556e-06,
144
+ "loss": 0.034,
145
+ "step": 900
146
+ },
147
+ {
148
+ "epoch": 13.194444444444445,
149
+ "grad_norm": 3.799290895462036,
150
+ "learning_rate": 9.004444444444445e-06,
151
+ "loss": 0.0407,
152
+ "step": 950
153
+ },
154
+ {
155
+ "epoch": 13.88888888888889,
156
+ "grad_norm": 3.571824789047241,
157
+ "learning_rate": 8.893333333333333e-06,
158
+ "loss": 0.0384,
159
+ "step": 1000
160
+ },
161
+ {
162
+ "epoch": 13.88888888888889,
163
+ "eval_emotion_accuracy": 0.4930555555555556,
164
+ "eval_loss": 0.19641591608524323,
165
+ "eval_runtime": 22.5447,
166
+ "eval_samples_per_second": 12.775,
167
+ "eval_steps_per_second": 1.597,
168
+ "eval_wer": 8.564814814814815,
169
+ "step": 1000
170
+ },
171
+ {
172
+ "epoch": 14.583333333333334,
173
+ "grad_norm": 10.1426362991333,
174
+ "learning_rate": 8.782222222222223e-06,
175
+ "loss": 0.0331,
176
+ "step": 1050
177
+ },
178
+ {
179
+ "epoch": 15.277777777777779,
180
+ "grad_norm": 3.474062204360962,
181
+ "learning_rate": 8.671111111111113e-06,
182
+ "loss": 0.0336,
183
+ "step": 1100
184
+ },
185
+ {
186
+ "epoch": 15.972222222222221,
187
+ "grad_norm": 5.525303363800049,
188
+ "learning_rate": 8.560000000000001e-06,
189
+ "loss": 0.0359,
190
+ "step": 1150
191
+ },
192
+ {
193
+ "epoch": 16.666666666666668,
194
+ "grad_norm": 4.540746212005615,
195
+ "learning_rate": 8.448888888888889e-06,
196
+ "loss": 0.0314,
197
+ "step": 1200
198
+ },
199
+ {
200
+ "epoch": 17.36111111111111,
201
+ "grad_norm": 3.207890033721924,
202
+ "learning_rate": 8.337777777777777e-06,
203
+ "loss": 0.0308,
204
+ "step": 1250
205
+ },
206
+ {
207
+ "epoch": 18.055555555555557,
208
+ "grad_norm": 7.226109981536865,
209
+ "learning_rate": 8.226666666666667e-06,
210
+ "loss": 0.0345,
211
+ "step": 1300
212
+ },
213
+ {
214
+ "epoch": 18.75,
215
+ "grad_norm": 6.696235179901123,
216
+ "learning_rate": 8.115555555555557e-06,
217
+ "loss": 0.0303,
218
+ "step": 1350
219
+ },
220
+ {
221
+ "epoch": 19.444444444444443,
222
+ "grad_norm": 7.980685710906982,
223
+ "learning_rate": 8.004444444444445e-06,
224
+ "loss": 0.0281,
225
+ "step": 1400
226
+ },
227
+ {
228
+ "epoch": 20.13888888888889,
229
+ "grad_norm": 5.168766498565674,
230
+ "learning_rate": 7.893333333333335e-06,
231
+ "loss": 0.0277,
232
+ "step": 1450
233
+ },
234
+ {
235
+ "epoch": 20.833333333333332,
236
+ "grad_norm": 5.401890754699707,
237
+ "learning_rate": 7.782222222222223e-06,
238
+ "loss": 0.0221,
239
+ "step": 1500
240
+ },
241
+ {
242
+ "epoch": 20.833333333333332,
243
+ "eval_emotion_accuracy": 0.4756944444444444,
244
+ "eval_loss": 0.2196139097213745,
245
+ "eval_runtime": 22.411,
246
+ "eval_samples_per_second": 12.851,
247
+ "eval_steps_per_second": 1.606,
248
+ "eval_wer": 8.854166666666668,
249
+ "step": 1500
250
+ },
251
+ {
252
+ "epoch": 21.52777777777778,
253
+ "grad_norm": 3.7461345195770264,
254
+ "learning_rate": 7.67111111111111e-06,
255
+ "loss": 0.0257,
256
+ "step": 1550
257
+ },
258
+ {
259
+ "epoch": 22.22222222222222,
260
+ "grad_norm": 5.136087894439697,
261
+ "learning_rate": 7.5600000000000005e-06,
262
+ "loss": 0.0256,
263
+ "step": 1600
264
+ },
265
+ {
266
+ "epoch": 22.916666666666668,
267
+ "grad_norm": 6.669163703918457,
268
+ "learning_rate": 7.44888888888889e-06,
269
+ "loss": 0.0247,
270
+ "step": 1650
271
+ },
272
+ {
273
+ "epoch": 23.61111111111111,
274
+ "grad_norm": 2.9139912128448486,
275
+ "learning_rate": 7.337777777777778e-06,
276
+ "loss": 0.0208,
277
+ "step": 1700
278
+ },
279
+ {
280
+ "epoch": 24.305555555555557,
281
+ "grad_norm": 5.5039777755737305,
282
+ "learning_rate": 7.226666666666667e-06,
283
+ "loss": 0.0223,
284
+ "step": 1750
285
+ },
286
+ {
287
+ "epoch": 25.0,
288
+ "grad_norm": 1.129085659980774,
289
+ "learning_rate": 7.115555555555557e-06,
290
+ "loss": 0.0193,
291
+ "step": 1800
292
+ },
293
+ {
294
+ "epoch": 25.694444444444443,
295
+ "grad_norm": 3.9426541328430176,
296
+ "learning_rate": 7.004444444444445e-06,
297
+ "loss": 0.0186,
298
+ "step": 1850
299
+ },
300
+ {
301
+ "epoch": 26.38888888888889,
302
+ "grad_norm": 2.649850845336914,
303
+ "learning_rate": 6.893333333333334e-06,
304
+ "loss": 0.0206,
305
+ "step": 1900
306
+ },
307
+ {
308
+ "epoch": 27.083333333333332,
309
+ "grad_norm": 4.883500576019287,
310
+ "learning_rate": 6.782222222222222e-06,
311
+ "loss": 0.0182,
312
+ "step": 1950
313
+ },
314
+ {
315
+ "epoch": 27.77777777777778,
316
+ "grad_norm": 6.350570201873779,
317
+ "learning_rate": 6.671111111111112e-06,
318
+ "loss": 0.0185,
319
+ "step": 2000
320
+ },
321
+ {
322
+ "epoch": 27.77777777777778,
323
+ "eval_emotion_accuracy": 0.5104166666666666,
324
+ "eval_loss": 0.21950353682041168,
325
+ "eval_runtime": 22.4407,
326
+ "eval_samples_per_second": 12.834,
327
+ "eval_steps_per_second": 1.604,
328
+ "eval_wer": 8.159722222222223,
329
+ "step": 2000
330
+ },
331
+ {
332
+ "epoch": 28.47222222222222,
333
+ "grad_norm": 4.34314489364624,
334
+ "learning_rate": 6.560000000000001e-06,
335
+ "loss": 0.0133,
336
+ "step": 2050
337
+ },
338
+ {
339
+ "epoch": 29.166666666666668,
340
+ "grad_norm": 1.1030272245407104,
341
+ "learning_rate": 6.448888888888889e-06,
342
+ "loss": 0.0165,
343
+ "step": 2100
344
+ },
345
+ {
346
+ "epoch": 29.86111111111111,
347
+ "grad_norm": 4.312460422515869,
348
+ "learning_rate": 6.3377777777777786e-06,
349
+ "loss": 0.0135,
350
+ "step": 2150
351
+ },
352
+ {
353
+ "epoch": 30.555555555555557,
354
+ "grad_norm": 1.9975934028625488,
355
+ "learning_rate": 6.2266666666666675e-06,
356
+ "loss": 0.0132,
357
+ "step": 2200
358
+ },
359
+ {
360
+ "epoch": 31.25,
361
+ "grad_norm": 0.5992574691772461,
362
+ "learning_rate": 6.1155555555555555e-06,
363
+ "loss": 0.0135,
364
+ "step": 2250
365
+ },
366
+ {
367
+ "epoch": 31.944444444444443,
368
+ "grad_norm": 2.138979434967041,
369
+ "learning_rate": 6.004444444444445e-06,
370
+ "loss": 0.0133,
371
+ "step": 2300
372
+ },
373
+ {
374
+ "epoch": 32.638888888888886,
375
+ "grad_norm": 2.354663372039795,
376
+ "learning_rate": 5.893333333333334e-06,
377
+ "loss": 0.0138,
378
+ "step": 2350
379
+ },
380
+ {
381
+ "epoch": 33.333333333333336,
382
+ "grad_norm": 5.00735330581665,
383
+ "learning_rate": 5.782222222222222e-06,
384
+ "loss": 0.0117,
385
+ "step": 2400
386
+ },
387
+ {
388
+ "epoch": 34.02777777777778,
389
+ "grad_norm": 1.3849042654037476,
390
+ "learning_rate": 5.671111111111112e-06,
391
+ "loss": 0.0111,
392
+ "step": 2450
393
+ },
394
+ {
395
+ "epoch": 34.72222222222222,
396
+ "grad_norm": 5.323485374450684,
397
+ "learning_rate": 5.560000000000001e-06,
398
+ "loss": 0.0113,
399
+ "step": 2500
400
+ },
401
+ {
402
+ "epoch": 34.72222222222222,
403
+ "eval_emotion_accuracy": 0.5,
404
+ "eval_loss": 0.24566501379013062,
405
+ "eval_runtime": 22.4775,
406
+ "eval_samples_per_second": 12.813,
407
+ "eval_steps_per_second": 1.602,
408
+ "eval_wer": 8.449074074074074,
409
+ "step": 2500
410
+ },
411
+ {
412
+ "epoch": 35.416666666666664,
413
+ "grad_norm": 0.9430980682373047,
414
+ "learning_rate": 5.448888888888889e-06,
415
+ "loss": 0.0099,
416
+ "step": 2550
417
+ },
418
+ {
419
+ "epoch": 36.111111111111114,
420
+ "grad_norm": 2.3679275512695312,
421
+ "learning_rate": 5.337777777777779e-06,
422
+ "loss": 0.0095,
423
+ "step": 2600
424
+ },
425
+ {
426
+ "epoch": 36.80555555555556,
427
+ "grad_norm": 1.6331442594528198,
428
+ "learning_rate": 5.226666666666667e-06,
429
+ "loss": 0.0106,
430
+ "step": 2650
431
+ },
432
+ {
433
+ "epoch": 37.5,
434
+ "grad_norm": 5.030411243438721,
435
+ "learning_rate": 5.115555555555556e-06,
436
+ "loss": 0.0096,
437
+ "step": 2700
438
+ },
439
+ {
440
+ "epoch": 38.19444444444444,
441
+ "grad_norm": 4.057655334472656,
442
+ "learning_rate": 5.004444444444445e-06,
443
+ "loss": 0.0086,
444
+ "step": 2750
445
+ },
446
+ {
447
+ "epoch": 38.888888888888886,
448
+ "grad_norm": 0.8502787947654724,
449
+ "learning_rate": 4.893333333333334e-06,
450
+ "loss": 0.0088,
451
+ "step": 2800
452
+ },
453
+ {
454
+ "epoch": 39.583333333333336,
455
+ "grad_norm": 0.8167718648910522,
456
+ "learning_rate": 4.784444444444445e-06,
457
+ "loss": 0.008,
458
+ "step": 2850
459
+ },
460
+ {
461
+ "epoch": 40.27777777777778,
462
+ "grad_norm": 2.7294929027557373,
463
+ "learning_rate": 4.673333333333333e-06,
464
+ "loss": 0.0075,
465
+ "step": 2900
466
+ },
467
+ {
468
+ "epoch": 40.97222222222222,
469
+ "grad_norm": 9.14332103729248,
470
+ "learning_rate": 4.562222222222222e-06,
471
+ "loss": 0.0066,
472
+ "step": 2950
473
+ },
474
+ {
475
+ "epoch": 41.666666666666664,
476
+ "grad_norm": 1.1356619596481323,
477
+ "learning_rate": 4.451111111111112e-06,
478
+ "loss": 0.0059,
479
+ "step": 3000
480
+ },
481
+ {
482
+ "epoch": 41.666666666666664,
483
+ "eval_emotion_accuracy": 0.53125,
484
+ "eval_loss": 0.24823080003261566,
485
+ "eval_runtime": 22.2933,
486
+ "eval_samples_per_second": 12.919,
487
+ "eval_steps_per_second": 1.615,
488
+ "eval_wer": 7.928240740740741,
489
+ "step": 3000
490
+ }
491
+ ],
492
+ "logging_steps": 50,
493
+ "max_steps": 5000,
494
+ "num_input_tokens_seen": 0,
495
+ "num_train_epochs": 70,
496
+ "save_steps": 500,
497
+ "stateful_callbacks": {
498
+ "TrainerControl": {
499
+ "args": {
500
+ "should_epoch_stop": false,
501
+ "should_evaluate": false,
502
+ "should_log": false,
503
+ "should_save": true,
504
+ "should_training_stop": false
505
+ },
506
+ "attributes": {}
507
+ }
508
+ },
509
+ "total_flos": 1.34330351616e+18,
510
+ "train_batch_size": 8,
511
+ "trial_name": null,
512
+ "trial_params": null
513
+ }
whisper-GST-exp2/checkpoint-3000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:250662ded9c17c3bffba67fe0499e745f5121205aa3494525fa64f50a2f65d02
3
+ size 5496
whisper-GST-exp2/checkpoint-5000/config.json ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "openai/whisper-tiny",
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": 384,
14
+ "decoder_attention_heads": 6,
15
+ "decoder_ffn_dim": 1536,
16
+ "decoder_layerdrop": 0.0,
17
+ "decoder_layers": 4,
18
+ "decoder_start_token_id": 50258,
19
+ "dropout": 0.0,
20
+ "encoder_attention_heads": 6,
21
+ "encoder_ffn_dim": 1536,
22
+ "encoder_layerdrop": 0.0,
23
+ "encoder_layers": 4,
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": 4,
53
+ "num_mel_bins": 80,
54
+ "pad_token_id": 50257,
55
+ "scale_embedding": false,
56
+ "torch_dtype": "float32",
57
+ "transformers_version": "4.48.0",
58
+ "use_cache": true,
59
+ "use_weighted_layer_sum": false,
60
+ "vocab_size": 51873
61
+ }
whisper-GST-exp2/checkpoint-5000/generation_config.json ADDED
@@ -0,0 +1,248 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alignment_heads": [
3
+ [
4
+ 2,
5
+ 2
6
+ ],
7
+ [
8
+ 3,
9
+ 0
10
+ ],
11
+ [
12
+ 3,
13
+ 2
14
+ ],
15
+ [
16
+ 3,
17
+ 3
18
+ ],
19
+ [
20
+ 3,
21
+ 4
22
+ ],
23
+ [
24
+ 3,
25
+ 5
26
+ ]
27
+ ],
28
+ "begin_suppress_tokens": [
29
+ 220,
30
+ 50257
31
+ ],
32
+ "bos_token_id": 50257,
33
+ "decoder_start_token_id": 50258,
34
+ "eos_token_id": 50257,
35
+ "forced_decoder_ids": [
36
+ [
37
+ 1,
38
+ null
39
+ ],
40
+ [
41
+ 2,
42
+ 50359
43
+ ]
44
+ ],
45
+ "is_multilingual": true,
46
+ "lang_to_id": {
47
+ "<|af|>": 50327,
48
+ "<|am|>": 50334,
49
+ "<|ar|>": 50272,
50
+ "<|as|>": 50350,
51
+ "<|az|>": 50304,
52
+ "<|ba|>": 50355,
53
+ "<|be|>": 50330,
54
+ "<|bg|>": 50292,
55
+ "<|bn|>": 50302,
56
+ "<|bo|>": 50347,
57
+ "<|br|>": 50309,
58
+ "<|bs|>": 50315,
59
+ "<|ca|>": 50270,
60
+ "<|cs|>": 50283,
61
+ "<|cy|>": 50297,
62
+ "<|da|>": 50285,
63
+ "<|de|>": 50261,
64
+ "<|el|>": 50281,
65
+ "<|en|>": 50259,
66
+ "<|es|>": 50262,
67
+ "<|et|>": 50307,
68
+ "<|eu|>": 50310,
69
+ "<|fa|>": 50300,
70
+ "<|fi|>": 50277,
71
+ "<|fo|>": 50338,
72
+ "<|fr|>": 50265,
73
+ "<|gl|>": 50319,
74
+ "<|gu|>": 50333,
75
+ "<|haw|>": 50352,
76
+ "<|ha|>": 50354,
77
+ "<|he|>": 50279,
78
+ "<|hi|>": 50276,
79
+ "<|hr|>": 50291,
80
+ "<|ht|>": 50339,
81
+ "<|hu|>": 50286,
82
+ "<|hy|>": 50312,
83
+ "<|id|>": 50275,
84
+ "<|is|>": 50311,
85
+ "<|it|>": 50274,
86
+ "<|ja|>": 50266,
87
+ "<|jw|>": 50356,
88
+ "<|ka|>": 50329,
89
+ "<|kk|>": 50316,
90
+ "<|km|>": 50323,
91
+ "<|kn|>": 50306,
92
+ "<|ko|>": 50264,
93
+ "<|la|>": 50294,
94
+ "<|lb|>": 50345,
95
+ "<|ln|>": 50353,
96
+ "<|lo|>": 50336,
97
+ "<|lt|>": 50293,
98
+ "<|lv|>": 50301,
99
+ "<|mg|>": 50349,
100
+ "<|mi|>": 50295,
101
+ "<|mk|>": 50308,
102
+ "<|ml|>": 50296,
103
+ "<|mn|>": 50314,
104
+ "<|mr|>": 50320,
105
+ "<|ms|>": 50282,
106
+ "<|mt|>": 50343,
107
+ "<|my|>": 50346,
108
+ "<|ne|>": 50313,
109
+ "<|nl|>": 50271,
110
+ "<|nn|>": 50342,
111
+ "<|no|>": 50288,
112
+ "<|oc|>": 50328,
113
+ "<|pa|>": 50321,
114
+ "<|pl|>": 50269,
115
+ "<|ps|>": 50340,
116
+ "<|pt|>": 50267,
117
+ "<|ro|>": 50284,
118
+ "<|ru|>": 50263,
119
+ "<|sa|>": 50344,
120
+ "<|sd|>": 50332,
121
+ "<|si|>": 50322,
122
+ "<|sk|>": 50298,
123
+ "<|sl|>": 50305,
124
+ "<|sn|>": 50324,
125
+ "<|so|>": 50326,
126
+ "<|sq|>": 50317,
127
+ "<|sr|>": 50303,
128
+ "<|su|>": 50357,
129
+ "<|sv|>": 50273,
130
+ "<|sw|>": 50318,
131
+ "<|ta|>": 50287,
132
+ "<|te|>": 50299,
133
+ "<|tg|>": 50331,
134
+ "<|th|>": 50289,
135
+ "<|tk|>": 50341,
136
+ "<|tl|>": 50348,
137
+ "<|tr|>": 50268,
138
+ "<|tt|>": 50351,
139
+ "<|uk|>": 50280,
140
+ "<|ur|>": 50290,
141
+ "<|uz|>": 50337,
142
+ "<|vi|>": 50278,
143
+ "<|yi|>": 50335,
144
+ "<|yo|>": 50325,
145
+ "<|zh|>": 50260
146
+ },
147
+ "max_initial_timestamp_index": 50,
148
+ "max_length": 448,
149
+ "no_timestamps_token_id": 50363,
150
+ "pad_token_id": 50257,
151
+ "prev_sot_token_id": 50361,
152
+ "return_timestamps": false,
153
+ "suppress_tokens": [
154
+ 1,
155
+ 2,
156
+ 7,
157
+ 8,
158
+ 9,
159
+ 10,
160
+ 14,
161
+ 25,
162
+ 26,
163
+ 27,
164
+ 28,
165
+ 29,
166
+ 31,
167
+ 58,
168
+ 59,
169
+ 60,
170
+ 61,
171
+ 62,
172
+ 63,
173
+ 90,
174
+ 91,
175
+ 92,
176
+ 93,
177
+ 359,
178
+ 503,
179
+ 522,
180
+ 542,
181
+ 873,
182
+ 893,
183
+ 902,
184
+ 918,
185
+ 922,
186
+ 931,
187
+ 1350,
188
+ 1853,
189
+ 1982,
190
+ 2460,
191
+ 2627,
192
+ 3246,
193
+ 3253,
194
+ 3268,
195
+ 3536,
196
+ 3846,
197
+ 3961,
198
+ 4183,
199
+ 4667,
200
+ 6585,
201
+ 6647,
202
+ 7273,
203
+ 9061,
204
+ 9383,
205
+ 10428,
206
+ 10929,
207
+ 11938,
208
+ 12033,
209
+ 12331,
210
+ 12562,
211
+ 13793,
212
+ 14157,
213
+ 14635,
214
+ 15265,
215
+ 15618,
216
+ 16553,
217
+ 16604,
218
+ 18362,
219
+ 18956,
220
+ 20075,
221
+ 21675,
222
+ 22520,
223
+ 26130,
224
+ 26161,
225
+ 26435,
226
+ 28279,
227
+ 29464,
228
+ 31650,
229
+ 32302,
230
+ 32470,
231
+ 36865,
232
+ 42863,
233
+ 47425,
234
+ 49870,
235
+ 50254,
236
+ 50258,
237
+ 50358,
238
+ 50359,
239
+ 50360,
240
+ 50361,
241
+ 50362
242
+ ],
243
+ "task_to_id": {
244
+ "transcribe": 50359,
245
+ "translate": 50358
246
+ },
247
+ "transformers_version": "4.48.0"
248
+ }
whisper-GST-exp2/checkpoint-5000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:650702af59d3ddedcab227f94f5d56dd090962a42cd3628b4aba902bde9d1844
3
+ size 160466696
whisper-GST-exp2/checkpoint-5000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b9f9168cec2ac0d4b267dc27c2fa18ad757a9cc11291f9fbe1416182379e89d1
3
+ size 321073018
whisper-GST-exp2/checkpoint-5000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f3d086aaf6277aa6d5780e6120b0c8c2209ef535924a39c4b2f143e69b7456db
3
+ size 14244
whisper-GST-exp2/checkpoint-5000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:01be5476e7a6e5111f1d5afbfb593f84a840b366b0ab9da29f51d3cd06049ede
3
+ size 1064
whisper-GST-exp2/checkpoint-5000/trainer_state.json ADDED
@@ -0,0 +1,833 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.53125,
3
+ "best_model_checkpoint": "./whisper-GST-exp2/checkpoint-3000",
4
+ "epoch": 69.44444444444444,
5
+ "eval_steps": 500,
6
+ "global_step": 5000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.6944444444444444,
13
+ "grad_norm": 5.084019660949707,
14
+ "learning_rate": 9.800000000000001e-07,
15
+ "loss": 0.0377,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 1.3888888888888888,
20
+ "grad_norm": 4.491146564483643,
21
+ "learning_rate": 1.98e-06,
22
+ "loss": 0.0402,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 2.0833333333333335,
27
+ "grad_norm": 6.822654724121094,
28
+ "learning_rate": 2.9800000000000003e-06,
29
+ "loss": 0.0411,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 2.7777777777777777,
34
+ "grad_norm": 2.353851318359375,
35
+ "learning_rate": 3.980000000000001e-06,
36
+ "loss": 0.0429,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 3.4722222222222223,
41
+ "grad_norm": 3.9458818435668945,
42
+ "learning_rate": 4.980000000000001e-06,
43
+ "loss": 0.0425,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 4.166666666666667,
48
+ "grad_norm": 6.142700672149658,
49
+ "learning_rate": 5.98e-06,
50
+ "loss": 0.045,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 4.861111111111111,
55
+ "grad_norm": 4.524245262145996,
56
+ "learning_rate": 6.98e-06,
57
+ "loss": 0.0455,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 5.555555555555555,
62
+ "grad_norm": 2.7861649990081787,
63
+ "learning_rate": 7.980000000000002e-06,
64
+ "loss": 0.0457,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 6.25,
69
+ "grad_norm": 5.412281513214111,
70
+ "learning_rate": 8.98e-06,
71
+ "loss": 0.045,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 6.944444444444445,
76
+ "grad_norm": 5.459417343139648,
77
+ "learning_rate": 9.960000000000001e-06,
78
+ "loss": 0.0502,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 6.944444444444445,
83
+ "eval_emotion_accuracy": 0.46875,
84
+ "eval_loss": 0.19364911317825317,
85
+ "eval_runtime": 22.6537,
86
+ "eval_samples_per_second": 12.713,
87
+ "eval_steps_per_second": 1.589,
88
+ "eval_wer": 8.969907407407407,
89
+ "step": 500
90
+ },
91
+ {
92
+ "epoch": 7.638888888888889,
93
+ "grad_norm": 3.5574755668640137,
94
+ "learning_rate": 9.893333333333334e-06,
95
+ "loss": 0.0469,
96
+ "step": 550
97
+ },
98
+ {
99
+ "epoch": 8.333333333333334,
100
+ "grad_norm": 4.717072010040283,
101
+ "learning_rate": 9.782222222222222e-06,
102
+ "loss": 0.0498,
103
+ "step": 600
104
+ },
105
+ {
106
+ "epoch": 9.027777777777779,
107
+ "grad_norm": 4.033562183380127,
108
+ "learning_rate": 9.671111111111112e-06,
109
+ "loss": 0.044,
110
+ "step": 650
111
+ },
112
+ {
113
+ "epoch": 9.722222222222221,
114
+ "grad_norm": 8.465327262878418,
115
+ "learning_rate": 9.56e-06,
116
+ "loss": 0.0434,
117
+ "step": 700
118
+ },
119
+ {
120
+ "epoch": 10.416666666666666,
121
+ "grad_norm": 3.5226399898529053,
122
+ "learning_rate": 9.44888888888889e-06,
123
+ "loss": 0.0358,
124
+ "step": 750
125
+ },
126
+ {
127
+ "epoch": 11.11111111111111,
128
+ "grad_norm": 1.9998486042022705,
129
+ "learning_rate": 9.33777777777778e-06,
130
+ "loss": 0.0421,
131
+ "step": 800
132
+ },
133
+ {
134
+ "epoch": 11.805555555555555,
135
+ "grad_norm": 2.3085238933563232,
136
+ "learning_rate": 9.226666666666668e-06,
137
+ "loss": 0.0412,
138
+ "step": 850
139
+ },
140
+ {
141
+ "epoch": 12.5,
142
+ "grad_norm": 2.7251176834106445,
143
+ "learning_rate": 9.115555555555556e-06,
144
+ "loss": 0.034,
145
+ "step": 900
146
+ },
147
+ {
148
+ "epoch": 13.194444444444445,
149
+ "grad_norm": 3.799290895462036,
150
+ "learning_rate": 9.004444444444445e-06,
151
+ "loss": 0.0407,
152
+ "step": 950
153
+ },
154
+ {
155
+ "epoch": 13.88888888888889,
156
+ "grad_norm": 3.571824789047241,
157
+ "learning_rate": 8.893333333333333e-06,
158
+ "loss": 0.0384,
159
+ "step": 1000
160
+ },
161
+ {
162
+ "epoch": 13.88888888888889,
163
+ "eval_emotion_accuracy": 0.4930555555555556,
164
+ "eval_loss": 0.19641591608524323,
165
+ "eval_runtime": 22.5447,
166
+ "eval_samples_per_second": 12.775,
167
+ "eval_steps_per_second": 1.597,
168
+ "eval_wer": 8.564814814814815,
169
+ "step": 1000
170
+ },
171
+ {
172
+ "epoch": 14.583333333333334,
173
+ "grad_norm": 10.1426362991333,
174
+ "learning_rate": 8.782222222222223e-06,
175
+ "loss": 0.0331,
176
+ "step": 1050
177
+ },
178
+ {
179
+ "epoch": 15.277777777777779,
180
+ "grad_norm": 3.474062204360962,
181
+ "learning_rate": 8.671111111111113e-06,
182
+ "loss": 0.0336,
183
+ "step": 1100
184
+ },
185
+ {
186
+ "epoch": 15.972222222222221,
187
+ "grad_norm": 5.525303363800049,
188
+ "learning_rate": 8.560000000000001e-06,
189
+ "loss": 0.0359,
190
+ "step": 1150
191
+ },
192
+ {
193
+ "epoch": 16.666666666666668,
194
+ "grad_norm": 4.540746212005615,
195
+ "learning_rate": 8.448888888888889e-06,
196
+ "loss": 0.0314,
197
+ "step": 1200
198
+ },
199
+ {
200
+ "epoch": 17.36111111111111,
201
+ "grad_norm": 3.207890033721924,
202
+ "learning_rate": 8.337777777777777e-06,
203
+ "loss": 0.0308,
204
+ "step": 1250
205
+ },
206
+ {
207
+ "epoch": 18.055555555555557,
208
+ "grad_norm": 7.226109981536865,
209
+ "learning_rate": 8.226666666666667e-06,
210
+ "loss": 0.0345,
211
+ "step": 1300
212
+ },
213
+ {
214
+ "epoch": 18.75,
215
+ "grad_norm": 6.696235179901123,
216
+ "learning_rate": 8.115555555555557e-06,
217
+ "loss": 0.0303,
218
+ "step": 1350
219
+ },
220
+ {
221
+ "epoch": 19.444444444444443,
222
+ "grad_norm": 7.980685710906982,
223
+ "learning_rate": 8.004444444444445e-06,
224
+ "loss": 0.0281,
225
+ "step": 1400
226
+ },
227
+ {
228
+ "epoch": 20.13888888888889,
229
+ "grad_norm": 5.168766498565674,
230
+ "learning_rate": 7.893333333333335e-06,
231
+ "loss": 0.0277,
232
+ "step": 1450
233
+ },
234
+ {
235
+ "epoch": 20.833333333333332,
236
+ "grad_norm": 5.401890754699707,
237
+ "learning_rate": 7.782222222222223e-06,
238
+ "loss": 0.0221,
239
+ "step": 1500
240
+ },
241
+ {
242
+ "epoch": 20.833333333333332,
243
+ "eval_emotion_accuracy": 0.4756944444444444,
244
+ "eval_loss": 0.2196139097213745,
245
+ "eval_runtime": 22.411,
246
+ "eval_samples_per_second": 12.851,
247
+ "eval_steps_per_second": 1.606,
248
+ "eval_wer": 8.854166666666668,
249
+ "step": 1500
250
+ },
251
+ {
252
+ "epoch": 21.52777777777778,
253
+ "grad_norm": 3.7461345195770264,
254
+ "learning_rate": 7.67111111111111e-06,
255
+ "loss": 0.0257,
256
+ "step": 1550
257
+ },
258
+ {
259
+ "epoch": 22.22222222222222,
260
+ "grad_norm": 5.136087894439697,
261
+ "learning_rate": 7.5600000000000005e-06,
262
+ "loss": 0.0256,
263
+ "step": 1600
264
+ },
265
+ {
266
+ "epoch": 22.916666666666668,
267
+ "grad_norm": 6.669163703918457,
268
+ "learning_rate": 7.44888888888889e-06,
269
+ "loss": 0.0247,
270
+ "step": 1650
271
+ },
272
+ {
273
+ "epoch": 23.61111111111111,
274
+ "grad_norm": 2.9139912128448486,
275
+ "learning_rate": 7.337777777777778e-06,
276
+ "loss": 0.0208,
277
+ "step": 1700
278
+ },
279
+ {
280
+ "epoch": 24.305555555555557,
281
+ "grad_norm": 5.5039777755737305,
282
+ "learning_rate": 7.226666666666667e-06,
283
+ "loss": 0.0223,
284
+ "step": 1750
285
+ },
286
+ {
287
+ "epoch": 25.0,
288
+ "grad_norm": 1.129085659980774,
289
+ "learning_rate": 7.115555555555557e-06,
290
+ "loss": 0.0193,
291
+ "step": 1800
292
+ },
293
+ {
294
+ "epoch": 25.694444444444443,
295
+ "grad_norm": 3.9426541328430176,
296
+ "learning_rate": 7.004444444444445e-06,
297
+ "loss": 0.0186,
298
+ "step": 1850
299
+ },
300
+ {
301
+ "epoch": 26.38888888888889,
302
+ "grad_norm": 2.649850845336914,
303
+ "learning_rate": 6.893333333333334e-06,
304
+ "loss": 0.0206,
305
+ "step": 1900
306
+ },
307
+ {
308
+ "epoch": 27.083333333333332,
309
+ "grad_norm": 4.883500576019287,
310
+ "learning_rate": 6.782222222222222e-06,
311
+ "loss": 0.0182,
312
+ "step": 1950
313
+ },
314
+ {
315
+ "epoch": 27.77777777777778,
316
+ "grad_norm": 6.350570201873779,
317
+ "learning_rate": 6.671111111111112e-06,
318
+ "loss": 0.0185,
319
+ "step": 2000
320
+ },
321
+ {
322
+ "epoch": 27.77777777777778,
323
+ "eval_emotion_accuracy": 0.5104166666666666,
324
+ "eval_loss": 0.21950353682041168,
325
+ "eval_runtime": 22.4407,
326
+ "eval_samples_per_second": 12.834,
327
+ "eval_steps_per_second": 1.604,
328
+ "eval_wer": 8.159722222222223,
329
+ "step": 2000
330
+ },
331
+ {
332
+ "epoch": 28.47222222222222,
333
+ "grad_norm": 4.34314489364624,
334
+ "learning_rate": 6.560000000000001e-06,
335
+ "loss": 0.0133,
336
+ "step": 2050
337
+ },
338
+ {
339
+ "epoch": 29.166666666666668,
340
+ "grad_norm": 1.1030272245407104,
341
+ "learning_rate": 6.448888888888889e-06,
342
+ "loss": 0.0165,
343
+ "step": 2100
344
+ },
345
+ {
346
+ "epoch": 29.86111111111111,
347
+ "grad_norm": 4.312460422515869,
348
+ "learning_rate": 6.3377777777777786e-06,
349
+ "loss": 0.0135,
350
+ "step": 2150
351
+ },
352
+ {
353
+ "epoch": 30.555555555555557,
354
+ "grad_norm": 1.9975934028625488,
355
+ "learning_rate": 6.2266666666666675e-06,
356
+ "loss": 0.0132,
357
+ "step": 2200
358
+ },
359
+ {
360
+ "epoch": 31.25,
361
+ "grad_norm": 0.5992574691772461,
362
+ "learning_rate": 6.1155555555555555e-06,
363
+ "loss": 0.0135,
364
+ "step": 2250
365
+ },
366
+ {
367
+ "epoch": 31.944444444444443,
368
+ "grad_norm": 2.138979434967041,
369
+ "learning_rate": 6.004444444444445e-06,
370
+ "loss": 0.0133,
371
+ "step": 2300
372
+ },
373
+ {
374
+ "epoch": 32.638888888888886,
375
+ "grad_norm": 2.354663372039795,
376
+ "learning_rate": 5.893333333333334e-06,
377
+ "loss": 0.0138,
378
+ "step": 2350
379
+ },
380
+ {
381
+ "epoch": 33.333333333333336,
382
+ "grad_norm": 5.00735330581665,
383
+ "learning_rate": 5.782222222222222e-06,
384
+ "loss": 0.0117,
385
+ "step": 2400
386
+ },
387
+ {
388
+ "epoch": 34.02777777777778,
389
+ "grad_norm": 1.3849042654037476,
390
+ "learning_rate": 5.671111111111112e-06,
391
+ "loss": 0.0111,
392
+ "step": 2450
393
+ },
394
+ {
395
+ "epoch": 34.72222222222222,
396
+ "grad_norm": 5.323485374450684,
397
+ "learning_rate": 5.560000000000001e-06,
398
+ "loss": 0.0113,
399
+ "step": 2500
400
+ },
401
+ {
402
+ "epoch": 34.72222222222222,
403
+ "eval_emotion_accuracy": 0.5,
404
+ "eval_loss": 0.24566501379013062,
405
+ "eval_runtime": 22.4775,
406
+ "eval_samples_per_second": 12.813,
407
+ "eval_steps_per_second": 1.602,
408
+ "eval_wer": 8.449074074074074,
409
+ "step": 2500
410
+ },
411
+ {
412
+ "epoch": 35.416666666666664,
413
+ "grad_norm": 0.9430980682373047,
414
+ "learning_rate": 5.448888888888889e-06,
415
+ "loss": 0.0099,
416
+ "step": 2550
417
+ },
418
+ {
419
+ "epoch": 36.111111111111114,
420
+ "grad_norm": 2.3679275512695312,
421
+ "learning_rate": 5.337777777777779e-06,
422
+ "loss": 0.0095,
423
+ "step": 2600
424
+ },
425
+ {
426
+ "epoch": 36.80555555555556,
427
+ "grad_norm": 1.6331442594528198,
428
+ "learning_rate": 5.226666666666667e-06,
429
+ "loss": 0.0106,
430
+ "step": 2650
431
+ },
432
+ {
433
+ "epoch": 37.5,
434
+ "grad_norm": 5.030411243438721,
435
+ "learning_rate": 5.115555555555556e-06,
436
+ "loss": 0.0096,
437
+ "step": 2700
438
+ },
439
+ {
440
+ "epoch": 38.19444444444444,
441
+ "grad_norm": 4.057655334472656,
442
+ "learning_rate": 5.004444444444445e-06,
443
+ "loss": 0.0086,
444
+ "step": 2750
445
+ },
446
+ {
447
+ "epoch": 38.888888888888886,
448
+ "grad_norm": 0.8502787947654724,
449
+ "learning_rate": 4.893333333333334e-06,
450
+ "loss": 0.0088,
451
+ "step": 2800
452
+ },
453
+ {
454
+ "epoch": 39.583333333333336,
455
+ "grad_norm": 0.8167718648910522,
456
+ "learning_rate": 4.784444444444445e-06,
457
+ "loss": 0.008,
458
+ "step": 2850
459
+ },
460
+ {
461
+ "epoch": 40.27777777777778,
462
+ "grad_norm": 2.7294929027557373,
463
+ "learning_rate": 4.673333333333333e-06,
464
+ "loss": 0.0075,
465
+ "step": 2900
466
+ },
467
+ {
468
+ "epoch": 40.97222222222222,
469
+ "grad_norm": 9.14332103729248,
470
+ "learning_rate": 4.562222222222222e-06,
471
+ "loss": 0.0066,
472
+ "step": 2950
473
+ },
474
+ {
475
+ "epoch": 41.666666666666664,
476
+ "grad_norm": 1.1356619596481323,
477
+ "learning_rate": 4.451111111111112e-06,
478
+ "loss": 0.0059,
479
+ "step": 3000
480
+ },
481
+ {
482
+ "epoch": 41.666666666666664,
483
+ "eval_emotion_accuracy": 0.53125,
484
+ "eval_loss": 0.24823080003261566,
485
+ "eval_runtime": 22.2933,
486
+ "eval_samples_per_second": 12.919,
487
+ "eval_steps_per_second": 1.615,
488
+ "eval_wer": 7.928240740740741,
489
+ "step": 3000
490
+ },
491
+ {
492
+ "epoch": 42.361111111111114,
493
+ "grad_norm": 3.7579808235168457,
494
+ "learning_rate": 4.34e-06,
495
+ "loss": 0.0054,
496
+ "step": 3050
497
+ },
498
+ {
499
+ "epoch": 43.05555555555556,
500
+ "grad_norm": 0.7577036023139954,
501
+ "learning_rate": 4.228888888888889e-06,
502
+ "loss": 0.0062,
503
+ "step": 3100
504
+ },
505
+ {
506
+ "epoch": 43.75,
507
+ "grad_norm": 1.215272307395935,
508
+ "learning_rate": 4.117777777777779e-06,
509
+ "loss": 0.0057,
510
+ "step": 3150
511
+ },
512
+ {
513
+ "epoch": 44.44444444444444,
514
+ "grad_norm": 0.528567910194397,
515
+ "learning_rate": 4.006666666666667e-06,
516
+ "loss": 0.0051,
517
+ "step": 3200
518
+ },
519
+ {
520
+ "epoch": 45.138888888888886,
521
+ "grad_norm": 1.1258264780044556,
522
+ "learning_rate": 3.895555555555556e-06,
523
+ "loss": 0.0061,
524
+ "step": 3250
525
+ },
526
+ {
527
+ "epoch": 45.833333333333336,
528
+ "grad_norm": 6.391242980957031,
529
+ "learning_rate": 3.784444444444445e-06,
530
+ "loss": 0.0054,
531
+ "step": 3300
532
+ },
533
+ {
534
+ "epoch": 46.52777777777778,
535
+ "grad_norm": 1.6513943672180176,
536
+ "learning_rate": 3.673333333333334e-06,
537
+ "loss": 0.0039,
538
+ "step": 3350
539
+ },
540
+ {
541
+ "epoch": 47.22222222222222,
542
+ "grad_norm": 0.24964100122451782,
543
+ "learning_rate": 3.5622222222222224e-06,
544
+ "loss": 0.0041,
545
+ "step": 3400
546
+ },
547
+ {
548
+ "epoch": 47.916666666666664,
549
+ "grad_norm": 0.6500294208526611,
550
+ "learning_rate": 3.4511111111111113e-06,
551
+ "loss": 0.0038,
552
+ "step": 3450
553
+ },
554
+ {
555
+ "epoch": 48.611111111111114,
556
+ "grad_norm": 0.19977302849292755,
557
+ "learning_rate": 3.3400000000000006e-06,
558
+ "loss": 0.0044,
559
+ "step": 3500
560
+ },
561
+ {
562
+ "epoch": 48.611111111111114,
563
+ "eval_emotion_accuracy": 0.5173611111111112,
564
+ "eval_loss": 0.25719186663627625,
565
+ "eval_runtime": 22.4845,
566
+ "eval_samples_per_second": 12.809,
567
+ "eval_steps_per_second": 1.601,
568
+ "eval_wer": 8.159722222222223,
569
+ "step": 3500
570
+ },
571
+ {
572
+ "epoch": 49.30555555555556,
573
+ "grad_norm": 0.22333833575248718,
574
+ "learning_rate": 3.228888888888889e-06,
575
+ "loss": 0.0036,
576
+ "step": 3550
577
+ },
578
+ {
579
+ "epoch": 50.0,
580
+ "grad_norm": 1.8754922151565552,
581
+ "learning_rate": 3.117777777777778e-06,
582
+ "loss": 0.0039,
583
+ "step": 3600
584
+ },
585
+ {
586
+ "epoch": 50.69444444444444,
587
+ "grad_norm": 1.2922059297561646,
588
+ "learning_rate": 3.0066666666666674e-06,
589
+ "loss": 0.0027,
590
+ "step": 3650
591
+ },
592
+ {
593
+ "epoch": 51.388888888888886,
594
+ "grad_norm": 0.8961352705955505,
595
+ "learning_rate": 2.895555555555556e-06,
596
+ "loss": 0.0036,
597
+ "step": 3700
598
+ },
599
+ {
600
+ "epoch": 52.083333333333336,
601
+ "grad_norm": 1.0021692514419556,
602
+ "learning_rate": 2.784444444444445e-06,
603
+ "loss": 0.0024,
604
+ "step": 3750
605
+ },
606
+ {
607
+ "epoch": 52.77777777777778,
608
+ "grad_norm": 1.5219653844833374,
609
+ "learning_rate": 2.6733333333333333e-06,
610
+ "loss": 0.0016,
611
+ "step": 3800
612
+ },
613
+ {
614
+ "epoch": 53.47222222222222,
615
+ "grad_norm": 0.991694450378418,
616
+ "learning_rate": 2.5622222222222226e-06,
617
+ "loss": 0.0017,
618
+ "step": 3850
619
+ },
620
+ {
621
+ "epoch": 54.166666666666664,
622
+ "grad_norm": 0.2559027671813965,
623
+ "learning_rate": 2.451111111111111e-06,
624
+ "loss": 0.0026,
625
+ "step": 3900
626
+ },
627
+ {
628
+ "epoch": 54.861111111111114,
629
+ "grad_norm": 1.1942185163497925,
630
+ "learning_rate": 2.3400000000000005e-06,
631
+ "loss": 0.0021,
632
+ "step": 3950
633
+ },
634
+ {
635
+ "epoch": 55.55555555555556,
636
+ "grad_norm": 1.747736930847168,
637
+ "learning_rate": 2.228888888888889e-06,
638
+ "loss": 0.0029,
639
+ "step": 4000
640
+ },
641
+ {
642
+ "epoch": 55.55555555555556,
643
+ "eval_emotion_accuracy": 0.5173611111111112,
644
+ "eval_loss": 0.26668617129325867,
645
+ "eval_runtime": 22.9511,
646
+ "eval_samples_per_second": 12.548,
647
+ "eval_steps_per_second": 1.569,
648
+ "eval_wer": 8.159722222222223,
649
+ "step": 4000
650
+ },
651
+ {
652
+ "epoch": 56.25,
653
+ "grad_norm": 1.9517264366149902,
654
+ "learning_rate": 2.117777777777778e-06,
655
+ "loss": 0.0031,
656
+ "step": 4050
657
+ },
658
+ {
659
+ "epoch": 56.94444444444444,
660
+ "grad_norm": 0.233982652425766,
661
+ "learning_rate": 2.006666666666667e-06,
662
+ "loss": 0.0019,
663
+ "step": 4100
664
+ },
665
+ {
666
+ "epoch": 57.638888888888886,
667
+ "grad_norm": 1.636906623840332,
668
+ "learning_rate": 1.8955555555555557e-06,
669
+ "loss": 0.0018,
670
+ "step": 4150
671
+ },
672
+ {
673
+ "epoch": 58.333333333333336,
674
+ "grad_norm": 0.18523189425468445,
675
+ "learning_rate": 1.7844444444444444e-06,
676
+ "loss": 0.0012,
677
+ "step": 4200
678
+ },
679
+ {
680
+ "epoch": 59.02777777777778,
681
+ "grad_norm": 0.19863753020763397,
682
+ "learning_rate": 1.6733333333333335e-06,
683
+ "loss": 0.0034,
684
+ "step": 4250
685
+ },
686
+ {
687
+ "epoch": 59.72222222222222,
688
+ "grad_norm": 1.4147911071777344,
689
+ "learning_rate": 1.5622222222222225e-06,
690
+ "loss": 0.0019,
691
+ "step": 4300
692
+ },
693
+ {
694
+ "epoch": 60.416666666666664,
695
+ "grad_norm": 0.24780753254890442,
696
+ "learning_rate": 1.4511111111111112e-06,
697
+ "loss": 0.002,
698
+ "step": 4350
699
+ },
700
+ {
701
+ "epoch": 61.111111111111114,
702
+ "grad_norm": 0.7788115739822388,
703
+ "learning_rate": 1.34e-06,
704
+ "loss": 0.0018,
705
+ "step": 4400
706
+ },
707
+ {
708
+ "epoch": 61.80555555555556,
709
+ "grad_norm": 2.4958019256591797,
710
+ "learning_rate": 1.228888888888889e-06,
711
+ "loss": 0.0014,
712
+ "step": 4450
713
+ },
714
+ {
715
+ "epoch": 62.5,
716
+ "grad_norm": 0.15582500398159027,
717
+ "learning_rate": 1.117777777777778e-06,
718
+ "loss": 0.0015,
719
+ "step": 4500
720
+ },
721
+ {
722
+ "epoch": 62.5,
723
+ "eval_emotion_accuracy": 0.5,
724
+ "eval_loss": 0.27334609627723694,
725
+ "eval_runtime": 22.6918,
726
+ "eval_samples_per_second": 12.692,
727
+ "eval_steps_per_second": 1.586,
728
+ "eval_wer": 8.449074074074074,
729
+ "step": 4500
730
+ },
731
+ {
732
+ "epoch": 63.19444444444444,
733
+ "grad_norm": 0.2193412482738495,
734
+ "learning_rate": 1.0066666666666668e-06,
735
+ "loss": 0.0011,
736
+ "step": 4550
737
+ },
738
+ {
739
+ "epoch": 63.888888888888886,
740
+ "grad_norm": 0.23676809668540955,
741
+ "learning_rate": 8.955555555555557e-07,
742
+ "loss": 0.0016,
743
+ "step": 4600
744
+ },
745
+ {
746
+ "epoch": 64.58333333333333,
747
+ "grad_norm": 10.430356979370117,
748
+ "learning_rate": 7.844444444444445e-07,
749
+ "loss": 0.0026,
750
+ "step": 4650
751
+ },
752
+ {
753
+ "epoch": 65.27777777777777,
754
+ "grad_norm": 0.498324990272522,
755
+ "learning_rate": 6.733333333333334e-07,
756
+ "loss": 0.0026,
757
+ "step": 4700
758
+ },
759
+ {
760
+ "epoch": 65.97222222222223,
761
+ "grad_norm": 1.9186415672302246,
762
+ "learning_rate": 5.622222222222223e-07,
763
+ "loss": 0.0009,
764
+ "step": 4750
765
+ },
766
+ {
767
+ "epoch": 66.66666666666667,
768
+ "grad_norm": 0.06345739960670471,
769
+ "learning_rate": 4.511111111111111e-07,
770
+ "loss": 0.0012,
771
+ "step": 4800
772
+ },
773
+ {
774
+ "epoch": 67.36111111111111,
775
+ "grad_norm": 0.705152153968811,
776
+ "learning_rate": 3.4000000000000003e-07,
777
+ "loss": 0.0013,
778
+ "step": 4850
779
+ },
780
+ {
781
+ "epoch": 68.05555555555556,
782
+ "grad_norm": 0.13428598642349243,
783
+ "learning_rate": 2.2888888888888892e-07,
784
+ "loss": 0.0015,
785
+ "step": 4900
786
+ },
787
+ {
788
+ "epoch": 68.75,
789
+ "grad_norm": 0.5085027813911438,
790
+ "learning_rate": 1.1777777777777778e-07,
791
+ "loss": 0.0013,
792
+ "step": 4950
793
+ },
794
+ {
795
+ "epoch": 69.44444444444444,
796
+ "grad_norm": 0.25888127088546753,
797
+ "learning_rate": 6.666666666666667e-09,
798
+ "loss": 0.0012,
799
+ "step": 5000
800
+ },
801
+ {
802
+ "epoch": 69.44444444444444,
803
+ "eval_emotion_accuracy": 0.5,
804
+ "eval_loss": 0.27628129720687866,
805
+ "eval_runtime": 22.618,
806
+ "eval_samples_per_second": 12.733,
807
+ "eval_steps_per_second": 1.592,
808
+ "eval_wer": 8.449074074074074,
809
+ "step": 5000
810
+ }
811
+ ],
812
+ "logging_steps": 50,
813
+ "max_steps": 5000,
814
+ "num_input_tokens_seen": 0,
815
+ "num_train_epochs": 70,
816
+ "save_steps": 500,
817
+ "stateful_callbacks": {
818
+ "TrainerControl": {
819
+ "args": {
820
+ "should_epoch_stop": false,
821
+ "should_evaluate": false,
822
+ "should_log": false,
823
+ "should_save": true,
824
+ "should_training_stop": true
825
+ },
826
+ "attributes": {}
827
+ }
828
+ },
829
+ "total_flos": 2.2388391936e+18,
830
+ "train_batch_size": 8,
831
+ "trial_name": null,
832
+ "trial_params": null
833
+ }
whisper-GST-exp2/checkpoint-5000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:250662ded9c17c3bffba67fe0499e745f5121205aa3494525fa64f50a2f65d02
3
+ size 5496