Hugging Face's logo

colerobertson
/
distilhubert-finetuned-not-a-word2

Audio Classification
Transformers
TensorBoard
Safetensors
hubert
Inference Endpoints
Model card Files Metrics Community
colerobertson commited on
Commit
a7fc36e
•
1 Parent(s): d89240d

Training in progress, epoch 1

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. model.safetensors +1 -1
  2. run-2/checkpoint-112/config.json +80 -0
  3. run-2/checkpoint-112/model.safetensors +3 -0
  4. run-2/checkpoint-112/optimizer.pt +3 -0
  5. run-2/checkpoint-112/preprocessor_config.json +9 -0
  6. run-2/checkpoint-112/rng_state.pth +3 -0
  7. run-2/checkpoint-112/scheduler.pt +3 -0
  8. run-2/checkpoint-112/trainer_state.json +241 -0
  9. run-2/checkpoint-112/training_args.bin +3 -0
  10. run-2/checkpoint-128/config.json +80 -0
  11. run-2/checkpoint-128/model.safetensors +3 -0
  12. run-2/checkpoint-128/optimizer.pt +3 -0
  13. run-2/checkpoint-128/preprocessor_config.json +9 -0
  14. run-2/checkpoint-128/rng_state.pth +3 -0
  15. run-2/checkpoint-128/scheduler.pt +3 -0
  16. run-2/checkpoint-128/trainer_state.json +271 -0
  17. run-2/checkpoint-128/training_args.bin +3 -0
  18. run-2/checkpoint-144/config.json +80 -0
  19. run-2/checkpoint-144/model.safetensors +3 -0
  20. run-2/checkpoint-144/optimizer.pt +3 -0
  21. run-2/checkpoint-144/preprocessor_config.json +9 -0
  22. run-2/checkpoint-144/rng_state.pth +3 -0
  23. run-2/checkpoint-144/scheduler.pt +3 -0
  24. run-2/checkpoint-144/trainer_state.json +301 -0
  25. run-2/checkpoint-144/training_args.bin +3 -0
  26. run-2/checkpoint-160/config.json +80 -0
  27. run-2/checkpoint-160/model.safetensors +3 -0
  28. run-2/checkpoint-160/optimizer.pt +3 -0
  29. run-2/checkpoint-160/preprocessor_config.json +9 -0
  30. run-2/checkpoint-160/rng_state.pth +3 -0
  31. run-2/checkpoint-160/scheduler.pt +3 -0
  32. run-2/checkpoint-160/trainer_state.json +338 -0
  33. run-2/checkpoint-160/training_args.bin +3 -0
  34. run-2/checkpoint-32/config.json +80 -0
  35. run-2/checkpoint-32/model.safetensors +3 -0
  36. run-2/checkpoint-32/optimizer.pt +3 -0
  37. run-2/checkpoint-32/preprocessor_config.json +9 -0
  38. run-2/checkpoint-32/rng_state.pth +3 -0
  39. run-2/checkpoint-32/scheduler.pt +3 -0
  40. run-2/checkpoint-32/trainer_state.json +84 -0
  41. run-2/checkpoint-32/training_args.bin +3 -0
  42. run-2/checkpoint-48/config.json +80 -0
  43. run-2/checkpoint-48/model.safetensors +3 -0
  44. run-2/checkpoint-48/optimizer.pt +3 -0
  45. run-2/checkpoint-48/preprocessor_config.json +9 -0
  46. run-2/checkpoint-48/rng_state.pth +3 -0
  47. run-2/checkpoint-48/scheduler.pt +3 -0
  48. run-2/checkpoint-48/trainer_state.json +114 -0
  49. run-2/checkpoint-48/training_args.bin +3 -0
  50. run-2/checkpoint-64/config.json +80 -0
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:42d09eafd2930cae4b789dc9b1dd21dab399f3369a7f7dab6297bd28e15031d7
3
  size 94763496
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:149f6b51cfaa089494700030336b7742cbe5c97fa8c55460d1626a4326b4e9e1
3
  size 94763496
run-2/checkpoint-112/config.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "ntu-spml/distilhubert",
3
+ "activation_dropout": 0.1,
4
+ "apply_spec_augment": false,
5
+ "architectures": [
6
+ "HubertForSequenceClassification"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 1,
10
+ "classifier_proj_size": 256,
11
+ "conv_bias": false,
12
+ "conv_dim": [
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512
20
+ ],
21
+ "conv_kernel": [
22
+ 10,
23
+ 3,
24
+ 3,
25
+ 3,
26
+ 3,
27
+ 2,
28
+ 2
29
+ ],
30
+ "conv_stride": [
31
+ 5,
32
+ 2,
33
+ 2,
34
+ 2,
35
+ 2,
36
+ 2,
37
+ 2
38
+ ],
39
+ "ctc_loss_reduction": "sum",
40
+ "ctc_zero_infinity": false,
41
+ "do_stable_layer_norm": false,
42
+ "eos_token_id": 2,
43
+ "feat_extract_activation": "gelu",
44
+ "feat_extract_norm": "group",
45
+ "feat_proj_dropout": 0.0,
46
+ "feat_proj_layer_norm": false,
47
+ "final_dropout": 0.0,
48
+ "hidden_act": "gelu",
49
+ "hidden_dropout": 0.1,
50
+ "hidden_size": 768,
51
+ "id2label": {
52
+ "0": "NOT_WORD",
53
+ "1": "WORD"
54
+ },
55
+ "initializer_range": 0.02,
56
+ "intermediate_size": 3072,
57
+ "label2id": {
58
+ "NOT_WORD": "0",
59
+ "WORD": "1"
60
+ },
61
+ "layer_norm_eps": 1e-05,
62
+ "layerdrop": 0.0,
63
+ "mask_feature_length": 10,
64
+ "mask_feature_min_masks": 0,
65
+ "mask_feature_prob": 0.0,
66
+ "mask_time_length": 10,
67
+ "mask_time_min_masks": 2,
68
+ "mask_time_prob": 0.05,
69
+ "model_type": "hubert",
70
+ "num_attention_heads": 12,
71
+ "num_conv_pos_embedding_groups": 16,
72
+ "num_conv_pos_embeddings": 128,
73
+ "num_feat_extract_layers": 7,
74
+ "num_hidden_layers": 2,
75
+ "pad_token_id": 0,
76
+ "torch_dtype": "float32",
77
+ "transformers_version": "4.38.1",
78
+ "use_weighted_layer_sum": false,
79
+ "vocab_size": 32
80
+ }
run-2/checkpoint-112/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8769f57e3fe659a8ade9250aa88b72970d213bebb97ea16c6801ff9746819013
3
+ size 94763496
run-2/checkpoint-112/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33ece6c2394683cd223b107c2a21760d9b52969d91fee883b846224b529dbdbe
3
+ size 189552570
run-2/checkpoint-112/preprocessor_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0,
7
+ "return_attention_mask": true,
8
+ "sampling_rate": 16000
9
+ }
run-2/checkpoint-112/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:36c809bb9b08d9afc629cd2084452ab3306bfc8750227d4673c218c6d5867e02
3
+ size 14244
run-2/checkpoint-112/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2524fddcbd328994dd6ff5d544548c6dc2002a878ff6793ebafcfd63ed62d286
3
+ size 1064
run-2/checkpoint-112/trainer_state.json ADDED
@@ -0,0 +1,241 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.7659574468085106,
3
+ "best_model_checkpoint": "distilhubert-finetuned-not-a-word2/run-2/checkpoint-112",
4
+ "epoch": 7.0,
5
+ "eval_steps": 500,
6
+ "global_step": 112,
7
+ "is_hyper_param_search": true,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.31,
13
+ "grad_norm": 0.8302626013755798,
14
+ "learning_rate": 9.517539799539297e-06,
15
+ "loss": 0.6966,
16
+ "step": 5
17
+ },
18
+ {
19
+ "epoch": 0.62,
20
+ "grad_norm": 0.6829280257225037,
21
+ "learning_rate": 1.9035079599078593e-05,
22
+ "loss": 0.6813,
23
+ "step": 10
24
+ },
25
+ {
26
+ "epoch": 0.94,
27
+ "grad_norm": 0.5496665835380554,
28
+ "learning_rate": 2.8552619398617893e-05,
29
+ "loss": 0.6451,
30
+ "step": 15
31
+ },
32
+ {
33
+ "epoch": 1.0,
34
+ "eval_f1": 0.7326732673267327,
35
+ "eval_loss": 0.6860618591308594,
36
+ "eval_runtime": 1.3401,
37
+ "eval_samples_per_second": 47.759,
38
+ "eval_steps_per_second": 5.97,
39
+ "step": 16
40
+ },
41
+ {
42
+ "epoch": 1.25,
43
+ "grad_norm": 1.3569624423980713,
44
+ "learning_rate": 2.9610123820788925e-05,
45
+ "loss": 0.6113,
46
+ "step": 20
47
+ },
48
+ {
49
+ "epoch": 1.56,
50
+ "grad_norm": 0.3838456869125366,
51
+ "learning_rate": 2.8552619398617893e-05,
52
+ "loss": 0.5997,
53
+ "step": 25
54
+ },
55
+ {
56
+ "epoch": 1.88,
57
+ "grad_norm": 0.46842828392982483,
58
+ "learning_rate": 2.7495114976446858e-05,
59
+ "loss": 0.6133,
60
+ "step": 30
61
+ },
62
+ {
63
+ "epoch": 2.0,
64
+ "eval_f1": 0.7326732673267327,
65
+ "eval_loss": 0.719573974609375,
66
+ "eval_runtime": 1.3426,
67
+ "eval_samples_per_second": 47.668,
68
+ "eval_steps_per_second": 5.958,
69
+ "step": 32
70
+ },
71
+ {
72
+ "epoch": 2.19,
73
+ "grad_norm": 0.7610273957252502,
74
+ "learning_rate": 2.6437610554275826e-05,
75
+ "loss": 0.5971,
76
+ "step": 35
77
+ },
78
+ {
79
+ "epoch": 2.5,
80
+ "grad_norm": 0.5527757406234741,
81
+ "learning_rate": 2.5380106132104794e-05,
82
+ "loss": 0.6373,
83
+ "step": 40
84
+ },
85
+ {
86
+ "epoch": 2.81,
87
+ "grad_norm": 0.9699375033378601,
88
+ "learning_rate": 2.4322601709933762e-05,
89
+ "loss": 0.5701,
90
+ "step": 45
91
+ },
92
+ {
93
+ "epoch": 3.0,
94
+ "eval_f1": 0.7326732673267327,
95
+ "eval_loss": 0.6987495422363281,
96
+ "eval_runtime": 1.344,
97
+ "eval_samples_per_second": 47.619,
98
+ "eval_steps_per_second": 5.952,
99
+ "step": 48
100
+ },
101
+ {
102
+ "epoch": 3.12,
103
+ "grad_norm": 0.5822675824165344,
104
+ "learning_rate": 2.3265097287762727e-05,
105
+ "loss": 0.5492,
106
+ "step": 50
107
+ },
108
+ {
109
+ "epoch": 3.44,
110
+ "grad_norm": 0.9859365224838257,
111
+ "learning_rate": 2.2207592865591692e-05,
112
+ "loss": 0.5552,
113
+ "step": 55
114
+ },
115
+ {
116
+ "epoch": 3.75,
117
+ "grad_norm": 0.7946156859397888,
118
+ "learning_rate": 2.115008844342066e-05,
119
+ "loss": 0.5459,
120
+ "step": 60
121
+ },
122
+ {
123
+ "epoch": 4.0,
124
+ "eval_f1": 0.7326732673267327,
125
+ "eval_loss": 0.6806888580322266,
126
+ "eval_runtime": 1.3463,
127
+ "eval_samples_per_second": 47.538,
128
+ "eval_steps_per_second": 5.942,
129
+ "step": 64
130
+ },
131
+ {
132
+ "epoch": 4.06,
133
+ "grad_norm": 1.2706314325332642,
134
+ "learning_rate": 2.009258402124963e-05,
135
+ "loss": 0.5792,
136
+ "step": 65
137
+ },
138
+ {
139
+ "epoch": 4.38,
140
+ "grad_norm": 1.456335186958313,
141
+ "learning_rate": 1.9035079599078593e-05,
142
+ "loss": 0.4967,
143
+ "step": 70
144
+ },
145
+ {
146
+ "epoch": 4.69,
147
+ "grad_norm": 1.6856915950775146,
148
+ "learning_rate": 1.797757517690756e-05,
149
+ "loss": 0.4966,
150
+ "step": 75
151
+ },
152
+ {
153
+ "epoch": 5.0,
154
+ "grad_norm": 4.080612659454346,
155
+ "learning_rate": 1.692007075473653e-05,
156
+ "loss": 0.4589,
157
+ "step": 80
158
+ },
159
+ {
160
+ "epoch": 5.0,
161
+ "eval_f1": 0.7474747474747475,
162
+ "eval_loss": 0.6495094299316406,
163
+ "eval_runtime": 1.348,
164
+ "eval_samples_per_second": 47.477,
165
+ "eval_steps_per_second": 5.935,
166
+ "step": 80
167
+ },
168
+ {
169
+ "epoch": 5.31,
170
+ "grad_norm": 2.8024661540985107,
171
+ "learning_rate": 1.5862566332565498e-05,
172
+ "loss": 0.4312,
173
+ "step": 85
174
+ },
175
+ {
176
+ "epoch": 5.62,
177
+ "grad_norm": 3.024564027786255,
178
+ "learning_rate": 1.4805061910394462e-05,
179
+ "loss": 0.4318,
180
+ "step": 90
181
+ },
182
+ {
183
+ "epoch": 5.94,
184
+ "grad_norm": 3.236055374145508,
185
+ "learning_rate": 1.3747557488223429e-05,
186
+ "loss": 0.3968,
187
+ "step": 95
188
+ },
189
+ {
190
+ "epoch": 6.0,
191
+ "eval_f1": 0.7391304347826088,
192
+ "eval_loss": 0.6395978927612305,
193
+ "eval_runtime": 1.3605,
194
+ "eval_samples_per_second": 47.043,
195
+ "eval_steps_per_second": 5.88,
196
+ "step": 96
197
+ },
198
+ {
199
+ "epoch": 6.25,
200
+ "grad_norm": 2.6648523807525635,
201
+ "learning_rate": 1.2690053066052397e-05,
202
+ "loss": 0.3623,
203
+ "step": 100
204
+ },
205
+ {
206
+ "epoch": 6.56,
207
+ "grad_norm": 2.6884357929229736,
208
+ "learning_rate": 1.1632548643881364e-05,
209
+ "loss": 0.4061,
210
+ "step": 105
211
+ },
212
+ {
213
+ "epoch": 6.88,
214
+ "grad_norm": 3.5860610008239746,
215
+ "learning_rate": 1.057504422171033e-05,
216
+ "loss": 0.3356,
217
+ "step": 110
218
+ },
219
+ {
220
+ "epoch": 7.0,
221
+ "eval_f1": 0.7659574468085106,
222
+ "eval_loss": 0.7118406295776367,
223
+ "eval_runtime": 1.3559,
224
+ "eval_samples_per_second": 47.202,
225
+ "eval_steps_per_second": 5.9,
226
+ "step": 112
227
+ }
228
+ ],
229
+ "logging_steps": 5,
230
+ "max_steps": 160,
231
+ "num_input_tokens_seen": 0,
232
+ "num_train_epochs": 10,
233
+ "save_steps": 500,
234
+ "total_flos": 6207811312717728.0,
235
+ "train_batch_size": 12,
236
+ "trial_name": null,
237
+ "trial_params": {
238
+ "learning_rate": 3.045612735852575e-05,
239
+ "per_device_train_batch_size": 12
240
+ }
241
+ }
run-2/checkpoint-112/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c88b324e8cf31191efb1727463a9a5c250d242da2e762cbea403fc8ef4c82b88
3
+ size 4920
run-2/checkpoint-128/config.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "ntu-spml/distilhubert",
3
+ "activation_dropout": 0.1,
4
+ "apply_spec_augment": false,
5
+ "architectures": [
6
+ "HubertForSequenceClassification"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 1,
10
+ "classifier_proj_size": 256,
11
+ "conv_bias": false,
12
+ "conv_dim": [
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512
20
+ ],
21
+ "conv_kernel": [
22
+ 10,
23
+ 3,
24
+ 3,
25
+ 3,
26
+ 3,
27
+ 2,
28
+ 2
29
+ ],
30
+ "conv_stride": [
31
+ 5,
32
+ 2,
33
+ 2,
34
+ 2,
35
+ 2,
36
+ 2,
37
+ 2
38
+ ],
39
+ "ctc_loss_reduction": "sum",
40
+ "ctc_zero_infinity": false,
41
+ "do_stable_layer_norm": false,
42
+ "eos_token_id": 2,
43
+ "feat_extract_activation": "gelu",
44
+ "feat_extract_norm": "group",
45
+ "feat_proj_dropout": 0.0,
46
+ "feat_proj_layer_norm": false,
47
+ "final_dropout": 0.0,
48
+ "hidden_act": "gelu",
49
+ "hidden_dropout": 0.1,
50
+ "hidden_size": 768,
51
+ "id2label": {
52
+ "0": "NOT_WORD",
53
+ "1": "WORD"
54
+ },
55
+ "initializer_range": 0.02,
56
+ "intermediate_size": 3072,
57
+ "label2id": {
58
+ "NOT_WORD": "0",
59
+ "WORD": "1"
60
+ },
61
+ "layer_norm_eps": 1e-05,
62
+ "layerdrop": 0.0,
63
+ "mask_feature_length": 10,
64
+ "mask_feature_min_masks": 0,
65
+ "mask_feature_prob": 0.0,
66
+ "mask_time_length": 10,
67
+ "mask_time_min_masks": 2,
68
+ "mask_time_prob": 0.05,
69
+ "model_type": "hubert",
70
+ "num_attention_heads": 12,
71
+ "num_conv_pos_embedding_groups": 16,
72
+ "num_conv_pos_embeddings": 128,
73
+ "num_feat_extract_layers": 7,
74
+ "num_hidden_layers": 2,
75
+ "pad_token_id": 0,
76
+ "torch_dtype": "float32",
77
+ "transformers_version": "4.38.1",
78
+ "use_weighted_layer_sum": false,
79
+ "vocab_size": 32
80
+ }
run-2/checkpoint-128/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9cb08b5d77235ea4890e7b575805ba8fcb9ec727262887873b7762c509aacc4a
3
+ size 94763496
run-2/checkpoint-128/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:49c0ffbbefcfa1e6a2e26f0f9b7760581a81aee9db943966b9187b22fa135f26
3
+ size 189552570
run-2/checkpoint-128/preprocessor_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0,
7
+ "return_attention_mask": true,
8
+ "sampling_rate": 16000
9
+ }
run-2/checkpoint-128/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed4e5381cc0c3a30927facf662967e2407924903702879ea24956b3087d4375d
3
+ size 14244
run-2/checkpoint-128/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:95adb274255733b1bd0af931726761303d3881cfcbadac2cddd3cb2e8318f5d5
3
+ size 1064
run-2/checkpoint-128/trainer_state.json ADDED
@@ -0,0 +1,271 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.7659574468085106,
3
+ "best_model_checkpoint": "distilhubert-finetuned-not-a-word2/run-2/checkpoint-112",
4
+ "epoch": 8.0,
5
+ "eval_steps": 500,
6
+ "global_step": 128,
7
+ "is_hyper_param_search": true,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.31,
13
+ "grad_norm": 0.8302626013755798,
14
+ "learning_rate": 9.517539799539297e-06,
15
+ "loss": 0.6966,
16
+ "step": 5
17
+ },
18
+ {
19
+ "epoch": 0.62,
20
+ "grad_norm": 0.6829280257225037,
21
+ "learning_rate": 1.9035079599078593e-05,
22
+ "loss": 0.6813,
23
+ "step": 10
24
+ },
25
+ {
26
+ "epoch": 0.94,
27
+ "grad_norm": 0.5496665835380554,
28
+ "learning_rate": 2.8552619398617893e-05,
29
+ "loss": 0.6451,
30
+ "step": 15
31
+ },
32
+ {
33
+ "epoch": 1.0,
34
+ "eval_f1": 0.7326732673267327,
35
+ "eval_loss": 0.6860618591308594,
36
+ "eval_runtime": 1.3401,
37
+ "eval_samples_per_second": 47.759,
38
+ "eval_steps_per_second": 5.97,
39
+ "step": 16
40
+ },
41
+ {
42
+ "epoch": 1.25,
43
+ "grad_norm": 1.3569624423980713,
44
+ "learning_rate": 2.9610123820788925e-05,
45
+ "loss": 0.6113,
46
+ "step": 20
47
+ },
48
+ {
49
+ "epoch": 1.56,
50
+ "grad_norm": 0.3838456869125366,
51
+ "learning_rate": 2.8552619398617893e-05,
52
+ "loss": 0.5997,
53
+ "step": 25
54
+ },
55
+ {
56
+ "epoch": 1.88,
57
+ "grad_norm": 0.46842828392982483,
58
+ "learning_rate": 2.7495114976446858e-05,
59
+ "loss": 0.6133,
60
+ "step": 30
61
+ },
62
+ {
63
+ "epoch": 2.0,
64
+ "eval_f1": 0.7326732673267327,
65
+ "eval_loss": 0.719573974609375,
66
+ "eval_runtime": 1.3426,
67
+ "eval_samples_per_second": 47.668,
68
+ "eval_steps_per_second": 5.958,
69
+ "step": 32
70
+ },
71
+ {
72
+ "epoch": 2.19,
73
+ "grad_norm": 0.7610273957252502,
74
+ "learning_rate": 2.6437610554275826e-05,
75
+ "loss": 0.5971,
76
+ "step": 35
77
+ },
78
+ {
79
+ "epoch": 2.5,
80
+ "grad_norm": 0.5527757406234741,
81
+ "learning_rate": 2.5380106132104794e-05,
82
+ "loss": 0.6373,
83
+ "step": 40
84
+ },
85
+ {
86
+ "epoch": 2.81,
87
+ "grad_norm": 0.9699375033378601,
88
+ "learning_rate": 2.4322601709933762e-05,
89
+ "loss": 0.5701,
90
+ "step": 45
91
+ },
92
+ {
93
+ "epoch": 3.0,
94
+ "eval_f1": 0.7326732673267327,
95
+ "eval_loss": 0.6987495422363281,
96
+ "eval_runtime": 1.344,
97
+ "eval_samples_per_second": 47.619,
98
+ "eval_steps_per_second": 5.952,
99
+ "step": 48
100
+ },
101
+ {
102
+ "epoch": 3.12,
103
+ "grad_norm": 0.5822675824165344,
104
+ "learning_rate": 2.3265097287762727e-05,
105
+ "loss": 0.5492,
106
+ "step": 50
107
+ },
108
+ {
109
+ "epoch": 3.44,
110
+ "grad_norm": 0.9859365224838257,
111
+ "learning_rate": 2.2207592865591692e-05,
112
+ "loss": 0.5552,
113
+ "step": 55
114
+ },
115
+ {
116
+ "epoch": 3.75,
117
+ "grad_norm": 0.7946156859397888,
118
+ "learning_rate": 2.115008844342066e-05,
119
+ "loss": 0.5459,
120
+ "step": 60
121
+ },
122
+ {
123
+ "epoch": 4.0,
124
+ "eval_f1": 0.7326732673267327,
125
+ "eval_loss": 0.6806888580322266,
126
+ "eval_runtime": 1.3463,
127
+ "eval_samples_per_second": 47.538,
128
+ "eval_steps_per_second": 5.942,
129
+ "step": 64
130
+ },
131
+ {
132
+ "epoch": 4.06,
133
+ "grad_norm": 1.2706314325332642,
134
+ "learning_rate": 2.009258402124963e-05,
135
+ "loss": 0.5792,
136
+ "step": 65
137
+ },
138
+ {
139
+ "epoch": 4.38,
140
+ "grad_norm": 1.456335186958313,
141
+ "learning_rate": 1.9035079599078593e-05,
142
+ "loss": 0.4967,
143
+ "step": 70
144
+ },
145
+ {
146
+ "epoch": 4.69,
147
+ "grad_norm": 1.6856915950775146,
148
+ "learning_rate": 1.797757517690756e-05,
149
+ "loss": 0.4966,
150
+ "step": 75
151
+ },
152
+ {
153
+ "epoch": 5.0,
154
+ "grad_norm": 4.080612659454346,
155
+ "learning_rate": 1.692007075473653e-05,
156
+ "loss": 0.4589,
157
+ "step": 80
158
+ },
159
+ {
160
+ "epoch": 5.0,
161
+ "eval_f1": 0.7474747474747475,
162
+ "eval_loss": 0.6495094299316406,
163
+ "eval_runtime": 1.348,
164
+ "eval_samples_per_second": 47.477,
165
+ "eval_steps_per_second": 5.935,
166
+ "step": 80
167
+ },
168
+ {
169
+ "epoch": 5.31,
170
+ "grad_norm": 2.8024661540985107,
171
+ "learning_rate": 1.5862566332565498e-05,
172
+ "loss": 0.4312,
173
+ "step": 85
174
+ },
175
+ {
176
+ "epoch": 5.62,
177
+ "grad_norm": 3.024564027786255,
178
+ "learning_rate": 1.4805061910394462e-05,
179
+ "loss": 0.4318,
180
+ "step": 90
181
+ },
182
+ {
183
+ "epoch": 5.94,
184
+ "grad_norm": 3.236055374145508,
185
+ "learning_rate": 1.3747557488223429e-05,
186
+ "loss": 0.3968,
187
+ "step": 95
188
+ },
189
+ {
190
+ "epoch": 6.0,
191
+ "eval_f1": 0.7391304347826088,
192
+ "eval_loss": 0.6395978927612305,
193
+ "eval_runtime": 1.3605,
194
+ "eval_samples_per_second": 47.043,
195
+ "eval_steps_per_second": 5.88,
196
+ "step": 96
197
+ },
198
+ {
199
+ "epoch": 6.25,
200
+ "grad_norm": 2.6648523807525635,
201
+ "learning_rate": 1.2690053066052397e-05,
202
+ "loss": 0.3623,
203
+ "step": 100
204
+ },
205
+ {
206
+ "epoch": 6.56,
207
+ "grad_norm": 2.6884357929229736,
208
+ "learning_rate": 1.1632548643881364e-05,
209
+ "loss": 0.4061,
210
+ "step": 105
211
+ },
212
+ {
213
+ "epoch": 6.88,
214
+ "grad_norm": 3.5860610008239746,
215
+ "learning_rate": 1.057504422171033e-05,
216
+ "loss": 0.3356,
217
+ "step": 110
218
+ },
219
+ {
220
+ "epoch": 7.0,
221
+ "eval_f1": 0.7659574468085106,
222
+ "eval_loss": 0.7118406295776367,
223
+ "eval_runtime": 1.3559,
224
+ "eval_samples_per_second": 47.202,
225
+ "eval_steps_per_second": 5.9,
226
+ "step": 112
227
+ },
228
+ {
229
+ "epoch": 7.19,
230
+ "grad_norm": 3.730743169784546,
231
+ "learning_rate": 9.517539799539297e-06,
232
+ "loss": 0.298,
233
+ "step": 115
234
+ },
235
+ {
236
+ "epoch": 7.5,
237
+ "grad_norm": 5.971029281616211,
238
+ "learning_rate": 8.460035377368265e-06,
239
+ "loss": 0.3183,
240
+ "step": 120
241
+ },
242
+ {
243
+ "epoch": 7.81,
244
+ "grad_norm": 3.3043053150177,
245
+ "learning_rate": 7.402530955197231e-06,
246
+ "loss": 0.2788,
247
+ "step": 125
248
+ },
249
+ {
250
+ "epoch": 8.0,
251
+ "eval_f1": 0.7272727272727273,
252
+ "eval_loss": 0.6860837936401367,
253
+ "eval_runtime": 1.3424,
254
+ "eval_samples_per_second": 47.676,
255
+ "eval_steps_per_second": 5.96,
256
+ "step": 128
257
+ }
258
+ ],
259
+ "logging_steps": 5,
260
+ "max_steps": 160,
261
+ "num_input_tokens_seen": 0,
262
+ "num_train_epochs": 10,
263
+ "save_steps": 500,
264
+ "total_flos": 7043982497129520.0,
265
+ "train_batch_size": 12,
266
+ "trial_name": null,
267
+ "trial_params": {
268
+ "learning_rate": 3.045612735852575e-05,
269
+ "per_device_train_batch_size": 12
270
+ }
271
+ }
run-2/checkpoint-128/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c88b324e8cf31191efb1727463a9a5c250d242da2e762cbea403fc8ef4c82b88
3
+ size 4920
run-2/checkpoint-144/config.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "ntu-spml/distilhubert",
3
+ "activation_dropout": 0.1,
4
+ "apply_spec_augment": false,
5
+ "architectures": [
6
+ "HubertForSequenceClassification"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 1,
10
+ "classifier_proj_size": 256,
11
+ "conv_bias": false,
12
+ "conv_dim": [
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512
20
+ ],
21
+ "conv_kernel": [
22
+ 10,
23
+ 3,
24
+ 3,
25
+ 3,
26
+ 3,
27
+ 2,
28
+ 2
29
+ ],
30
+ "conv_stride": [
31
+ 5,
32
+ 2,
33
+ 2,
34
+ 2,
35
+ 2,
36
+ 2,
37
+ 2
38
+ ],
39
+ "ctc_loss_reduction": "sum",
40
+ "ctc_zero_infinity": false,
41
+ "do_stable_layer_norm": false,
42
+ "eos_token_id": 2,
43
+ "feat_extract_activation": "gelu",
44
+ "feat_extract_norm": "group",
45
+ "feat_proj_dropout": 0.0,
46
+ "feat_proj_layer_norm": false,
47
+ "final_dropout": 0.0,
48
+ "hidden_act": "gelu",
49
+ "hidden_dropout": 0.1,
50
+ "hidden_size": 768,
51
+ "id2label": {
52
+ "0": "NOT_WORD",
53
+ "1": "WORD"
54
+ },
55
+ "initializer_range": 0.02,
56
+ "intermediate_size": 3072,
57
+ "label2id": {
58
+ "NOT_WORD": "0",
59
+ "WORD": "1"
60
+ },
61
+ "layer_norm_eps": 1e-05,
62
+ "layerdrop": 0.0,
63
+ "mask_feature_length": 10,
64
+ "mask_feature_min_masks": 0,
65
+ "mask_feature_prob": 0.0,
66
+ "mask_time_length": 10,
67
+ "mask_time_min_masks": 2,
68
+ "mask_time_prob": 0.05,
69
+ "model_type": "hubert",
70
+ "num_attention_heads": 12,
71
+ "num_conv_pos_embedding_groups": 16,
72
+ "num_conv_pos_embeddings": 128,
73
+ "num_feat_extract_layers": 7,
74
+ "num_hidden_layers": 2,
75
+ "pad_token_id": 0,
76
+ "torch_dtype": "float32",
77
+ "transformers_version": "4.38.1",
78
+ "use_weighted_layer_sum": false,
79
+ "vocab_size": 32
80
+ }
run-2/checkpoint-144/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4d17b33be18635d847b67dd0ca131c83952c6e80bb65f970cf6d8e53bce18d15
3
+ size 94763496
run-2/checkpoint-144/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:640312979bd816fe24015c38a34313818d20cb4b8670366969657dbbd68e62a2
3
+ size 189552570
run-2/checkpoint-144/preprocessor_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0,
7
+ "return_attention_mask": true,
8
+ "sampling_rate": 16000
9
+ }
run-2/checkpoint-144/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a676191a9226b196013b0001ff6f13f8b83d2ce7b0a01a243ac19bd9905efdc8
3
+ size 14244
run-2/checkpoint-144/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f56bd50cb83e7e4aea9e61f658cc3774f5f196fc44e20020d5d70039f68429e3
3
+ size 1064
run-2/checkpoint-144/trainer_state.json ADDED
@@ -0,0 +1,301 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.7659574468085106,
3
+ "best_model_checkpoint": "distilhubert-finetuned-not-a-word2/run-2/checkpoint-112",
4
+ "epoch": 9.0,
5
+ "eval_steps": 500,
6
+ "global_step": 144,
7
+ "is_hyper_param_search": true,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.31,
13
+ "grad_norm": 0.8302626013755798,
14
+ "learning_rate": 9.517539799539297e-06,
15
+ "loss": 0.6966,
16
+ "step": 5
17
+ },
18
+ {
19
+ "epoch": 0.62,
20
+ "grad_norm": 0.6829280257225037,
21
+ "learning_rate": 1.9035079599078593e-05,
22
+ "loss": 0.6813,
23
+ "step": 10
24
+ },
25
+ {
26
+ "epoch": 0.94,
27
+ "grad_norm": 0.5496665835380554,
28
+ "learning_rate": 2.8552619398617893e-05,
29
+ "loss": 0.6451,
30
+ "step": 15
31
+ },
32
+ {
33
+ "epoch": 1.0,
34
+ "eval_f1": 0.7326732673267327,
35
+ "eval_loss": 0.6860618591308594,
36
+ "eval_runtime": 1.3401,
37
+ "eval_samples_per_second": 47.759,
38
+ "eval_steps_per_second": 5.97,
39
+ "step": 16
40
+ },
41
+ {
42
+ "epoch": 1.25,
43
+ "grad_norm": 1.3569624423980713,
44
+ "learning_rate": 2.9610123820788925e-05,
45
+ "loss": 0.6113,
46
+ "step": 20
47
+ },
48
+ {
49
+ "epoch": 1.56,
50
+ "grad_norm": 0.3838456869125366,
51
+ "learning_rate": 2.8552619398617893e-05,
52
+ "loss": 0.5997,
53
+ "step": 25
54
+ },
55
+ {
56
+ "epoch": 1.88,
57
+ "grad_norm": 0.46842828392982483,
58
+ "learning_rate": 2.7495114976446858e-05,
59
+ "loss": 0.6133,
60
+ "step": 30
61
+ },
62
+ {
63
+ "epoch": 2.0,
64
+ "eval_f1": 0.7326732673267327,
65
+ "eval_loss": 0.719573974609375,
66
+ "eval_runtime": 1.3426,
67
+ "eval_samples_per_second": 47.668,
68
+ "eval_steps_per_second": 5.958,
69
+ "step": 32
70
+ },
71
+ {
72
+ "epoch": 2.19,
73
+ "grad_norm": 0.7610273957252502,
74
+ "learning_rate": 2.6437610554275826e-05,
75
+ "loss": 0.5971,
76
+ "step": 35
77
+ },
78
+ {
79
+ "epoch": 2.5,
80
+ "grad_norm": 0.5527757406234741,
81
+ "learning_rate": 2.5380106132104794e-05,
82
+ "loss": 0.6373,
83
+ "step": 40
84
+ },
85
+ {
86
+ "epoch": 2.81,
87
+ "grad_norm": 0.9699375033378601,
88
+ "learning_rate": 2.4322601709933762e-05,
89
+ "loss": 0.5701,
90
+ "step": 45
91
+ },
92
+ {
93
+ "epoch": 3.0,
94
+ "eval_f1": 0.7326732673267327,
95
+ "eval_loss": 0.6987495422363281,
96
+ "eval_runtime": 1.344,
97
+ "eval_samples_per_second": 47.619,
98
+ "eval_steps_per_second": 5.952,
99
+ "step": 48
100
+ },
101
+ {
102
+ "epoch": 3.12,
103
+ "grad_norm": 0.5822675824165344,
104
+ "learning_rate": 2.3265097287762727e-05,
105
+ "loss": 0.5492,
106
+ "step": 50
107
+ },
108
+ {
109
+ "epoch": 3.44,
110
+ "grad_norm": 0.9859365224838257,
111
+ "learning_rate": 2.2207592865591692e-05,
112
+ "loss": 0.5552,
113
+ "step": 55
114
+ },
115
+ {
116
+ "epoch": 3.75,
117
+ "grad_norm": 0.7946156859397888,
118
+ "learning_rate": 2.115008844342066e-05,
119
+ "loss": 0.5459,
120
+ "step": 60
121
+ },
122
+ {
123
+ "epoch": 4.0,
124
+ "eval_f1": 0.7326732673267327,
125
+ "eval_loss": 0.6806888580322266,
126
+ "eval_runtime": 1.3463,
127
+ "eval_samples_per_second": 47.538,
128
+ "eval_steps_per_second": 5.942,
129
+ "step": 64
130
+ },
131
+ {
132
+ "epoch": 4.06,
133
+ "grad_norm": 1.2706314325332642,
134
+ "learning_rate": 2.009258402124963e-05,
135
+ "loss": 0.5792,
136
+ "step": 65
137
+ },
138
+ {
139
+ "epoch": 4.38,
140
+ "grad_norm": 1.456335186958313,
141
+ "learning_rate": 1.9035079599078593e-05,
142
+ "loss": 0.4967,
143
+ "step": 70
144
+ },
145
+ {
146
+ "epoch": 4.69,
147
+ "grad_norm": 1.6856915950775146,
148
+ "learning_rate": 1.797757517690756e-05,
149
+ "loss": 0.4966,
150
+ "step": 75
151
+ },
152
+ {
153
+ "epoch": 5.0,
154
+ "grad_norm": 4.080612659454346,
155
+ "learning_rate": 1.692007075473653e-05,
156
+ "loss": 0.4589,
157
+ "step": 80
158
+ },
159
+ {
160
+ "epoch": 5.0,
161
+ "eval_f1": 0.7474747474747475,
162
+ "eval_loss": 0.6495094299316406,
163
+ "eval_runtime": 1.348,
164
+ "eval_samples_per_second": 47.477,
165
+ "eval_steps_per_second": 5.935,
166
+ "step": 80
167
+ },
168
+ {
169
+ "epoch": 5.31,
170
+ "grad_norm": 2.8024661540985107,
171
+ "learning_rate": 1.5862566332565498e-05,
172
+ "loss": 0.4312,
173
+ "step": 85
174
+ },
175
+ {
176
+ "epoch": 5.62,
177
+ "grad_norm": 3.024564027786255,
178
+ "learning_rate": 1.4805061910394462e-05,
179
+ "loss": 0.4318,
180
+ "step": 90
181
+ },
182
+ {
183
+ "epoch": 5.94,
184
+ "grad_norm": 3.236055374145508,
185
+ "learning_rate": 1.3747557488223429e-05,
186
+ "loss": 0.3968,
187
+ "step": 95
188
+ },
189
+ {
190
+ "epoch": 6.0,
191
+ "eval_f1": 0.7391304347826088,
192
+ "eval_loss": 0.6395978927612305,
193
+ "eval_runtime": 1.3605,
194
+ "eval_samples_per_second": 47.043,
195
+ "eval_steps_per_second": 5.88,
196
+ "step": 96
197
+ },
198
+ {
199
+ "epoch": 6.25,
200
+ "grad_norm": 2.6648523807525635,
201
+ "learning_rate": 1.2690053066052397e-05,
202
+ "loss": 0.3623,
203
+ "step": 100
204
+ },
205
+ {
206
+ "epoch": 6.56,
207
+ "grad_norm": 2.6884357929229736,
208
+ "learning_rate": 1.1632548643881364e-05,
209
+ "loss": 0.4061,
210
+ "step": 105
211
+ },
212
+ {
213
+ "epoch": 6.88,
214
+ "grad_norm": 3.5860610008239746,
215
+ "learning_rate": 1.057504422171033e-05,
216
+ "loss": 0.3356,
217
+ "step": 110
218
+ },
219
+ {
220
+ "epoch": 7.0,
221
+ "eval_f1": 0.7659574468085106,
222
+ "eval_loss": 0.7118406295776367,
223
+ "eval_runtime": 1.3559,
224
+ "eval_samples_per_second": 47.202,
225
+ "eval_steps_per_second": 5.9,
226
+ "step": 112
227
+ },
228
+ {
229
+ "epoch": 7.19,
230
+ "grad_norm": 3.730743169784546,
231
+ "learning_rate": 9.517539799539297e-06,
232
+ "loss": 0.298,
233
+ "step": 115
234
+ },
235
+ {
236
+ "epoch": 7.5,
237
+ "grad_norm": 5.971029281616211,
238
+ "learning_rate": 8.460035377368265e-06,
239
+ "loss": 0.3183,
240
+ "step": 120
241
+ },
242
+ {
243
+ "epoch": 7.81,
244
+ "grad_norm": 3.3043053150177,
245
+ "learning_rate": 7.402530955197231e-06,
246
+ "loss": 0.2788,
247
+ "step": 125
248
+ },
249
+ {
250
+ "epoch": 8.0,
251
+ "eval_f1": 0.7272727272727273,
252
+ "eval_loss": 0.6860837936401367,
253
+ "eval_runtime": 1.3424,
254
+ "eval_samples_per_second": 47.676,
255
+ "eval_steps_per_second": 5.96,
256
+ "step": 128
257
+ },
258
+ {
259
+ "epoch": 8.12,
260
+ "grad_norm": 3.310666084289551,
261
+ "learning_rate": 6.556527417460405e-06,
262
+ "loss": 0.3052,
263
+ "step": 130
264
+ },
265
+ {
266
+ "epoch": 8.44,
267
+ "grad_norm": 3.9955413341522217,
268
+ "learning_rate": 5.499022995289372e-06,
269
+ "loss": 0.2833,
270
+ "step": 135
271
+ },
272
+ {
273
+ "epoch": 8.75,
274
+ "grad_norm": 2.4762678146362305,
275
+ "learning_rate": 4.441518573118339e-06,
276
+ "loss": 0.2387,
277
+ "step": 140
278
+ },
279
+ {
280
+ "epoch": 9.0,
281
+ "eval_f1": 0.7333333333333333,
282
+ "eval_loss": 0.7196941375732422,
283
+ "eval_runtime": 1.3575,
284
+ "eval_samples_per_second": 47.147,
285
+ "eval_steps_per_second": 5.893,
286
+ "step": 144
287
+ }
288
+ ],
289
+ "logging_steps": 5,
290
+ "max_steps": 160,
291
+ "num_input_tokens_seen": 0,
292
+ "num_train_epochs": 10,
293
+ "save_steps": 500,
294
+ "total_flos": 7846324158364704.0,
295
+ "train_batch_size": 12,
296
+ "trial_name": null,
297
+ "trial_params": {
298
+ "learning_rate": 3.045612735852575e-05,
299
+ "per_device_train_batch_size": 12
300
+ }
301
+ }
run-2/checkpoint-144/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c88b324e8cf31191efb1727463a9a5c250d242da2e762cbea403fc8ef4c82b88
3
+ size 4920
run-2/checkpoint-160/config.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "ntu-spml/distilhubert",
3
+ "activation_dropout": 0.1,
4
+ "apply_spec_augment": false,
5
+ "architectures": [
6
+ "HubertForSequenceClassification"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 1,
10
+ "classifier_proj_size": 256,
11
+ "conv_bias": false,
12
+ "conv_dim": [
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512
20
+ ],
21
+ "conv_kernel": [
22
+ 10,
23
+ 3,
24
+ 3,
25
+ 3,
26
+ 3,
27
+ 2,
28
+ 2
29
+ ],
30
+ "conv_stride": [
31
+ 5,
32
+ 2,
33
+ 2,
34
+ 2,
35
+ 2,
36
+ 2,
37
+ 2
38
+ ],
39
+ "ctc_loss_reduction": "sum",
40
+ "ctc_zero_infinity": false,
41
+ "do_stable_layer_norm": false,
42
+ "eos_token_id": 2,
43
+ "feat_extract_activation": "gelu",
44
+ "feat_extract_norm": "group",
45
+ "feat_proj_dropout": 0.0,
46
+ "feat_proj_layer_norm": false,
47
+ "final_dropout": 0.0,
48
+ "hidden_act": "gelu",
49
+ "hidden_dropout": 0.1,
50
+ "hidden_size": 768,
51
+ "id2label": {
52
+ "0": "NOT_WORD",
53
+ "1": "WORD"
54
+ },
55
+ "initializer_range": 0.02,
56
+ "intermediate_size": 3072,
57
+ "label2id": {
58
+ "NOT_WORD": "0",
59
+ "WORD": "1"
60
+ },
61
+ "layer_norm_eps": 1e-05,
62
+ "layerdrop": 0.0,
63
+ "mask_feature_length": 10,
64
+ "mask_feature_min_masks": 0,
65
+ "mask_feature_prob": 0.0,
66
+ "mask_time_length": 10,
67
+ "mask_time_min_masks": 2,
68
+ "mask_time_prob": 0.05,
69
+ "model_type": "hubert",
70
+ "num_attention_heads": 12,
71
+ "num_conv_pos_embedding_groups": 16,
72
+ "num_conv_pos_embeddings": 128,
73
+ "num_feat_extract_layers": 7,
74
+ "num_hidden_layers": 2,
75
+ "pad_token_id": 0,
76
+ "torch_dtype": "float32",
77
+ "transformers_version": "4.38.1",
78
+ "use_weighted_layer_sum": false,
79
+ "vocab_size": 32
80
+ }
run-2/checkpoint-160/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3dca1c25b7cc0d11f317b92bdc8ee9702bd79e60aebc9d807dc1929e58030af2
3
+ size 94763496
run-2/checkpoint-160/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d82b25f9adc9d5907157ec89ace2ea7eac835f20d4fbf266f8721de0dd72c88b
3
+ size 189552570
run-2/checkpoint-160/preprocessor_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0,
7
+ "return_attention_mask": true,
8
+ "sampling_rate": 16000
9
+ }
run-2/checkpoint-160/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e90298b4a8c0d0dbda066de5acda250b10d21809928797d74a473153181aa7f4
3
+ size 14244
run-2/checkpoint-160/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7cf64dca30d41ccc1f1b40ba88c9c550d85c4fd9b610d54a30b4016ef5b0c837
3
+ size 1064
run-2/checkpoint-160/trainer_state.json ADDED
@@ -0,0 +1,338 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.7659574468085106,
3
+ "best_model_checkpoint": "distilhubert-finetuned-not-a-word2/run-2/checkpoint-112",
4
+ "epoch": 10.0,
5
+ "eval_steps": 500,
6
+ "global_step": 160,
7
+ "is_hyper_param_search": true,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.31,
13
+ "grad_norm": 0.8302626013755798,
14
+ "learning_rate": 9.517539799539297e-06,
15
+ "loss": 0.6966,
16
+ "step": 5
17
+ },
18
+ {
19
+ "epoch": 0.62,
20
+ "grad_norm": 0.6829280257225037,
21
+ "learning_rate": 1.9035079599078593e-05,
22
+ "loss": 0.6813,
23
+ "step": 10
24
+ },
25
+ {
26
+ "epoch": 0.94,
27
+ "grad_norm": 0.5496665835380554,
28
+ "learning_rate": 2.8552619398617893e-05,
29
+ "loss": 0.6451,
30
+ "step": 15
31
+ },
32
+ {
33
+ "epoch": 1.0,
34
+ "eval_f1": 0.7326732673267327,
35
+ "eval_loss": 0.6860618591308594,
36
+ "eval_runtime": 1.3401,
37
+ "eval_samples_per_second": 47.759,
38
+ "eval_steps_per_second": 5.97,
39
+ "step": 16
40
+ },
41
+ {
42
+ "epoch": 1.25,
43
+ "grad_norm": 1.3569624423980713,
44
+ "learning_rate": 2.9610123820788925e-05,
45
+ "loss": 0.6113,
46
+ "step": 20
47
+ },
48
+ {
49
+ "epoch": 1.56,
50
+ "grad_norm": 0.3838456869125366,
51
+ "learning_rate": 2.8552619398617893e-05,
52
+ "loss": 0.5997,
53
+ "step": 25
54
+ },
55
+ {
56
+ "epoch": 1.88,
57
+ "grad_norm": 0.46842828392982483,
58
+ "learning_rate": 2.7495114976446858e-05,
59
+ "loss": 0.6133,
60
+ "step": 30
61
+ },
62
+ {
63
+ "epoch": 2.0,
64
+ "eval_f1": 0.7326732673267327,
65
+ "eval_loss": 0.719573974609375,
66
+ "eval_runtime": 1.3426,
67
+ "eval_samples_per_second": 47.668,
68
+ "eval_steps_per_second": 5.958,
69
+ "step": 32
70
+ },
71
+ {
72
+ "epoch": 2.19,
73
+ "grad_norm": 0.7610273957252502,
74
+ "learning_rate": 2.6437610554275826e-05,
75
+ "loss": 0.5971,
76
+ "step": 35
77
+ },
78
+ {
79
+ "epoch": 2.5,
80
+ "grad_norm": 0.5527757406234741,
81
+ "learning_rate": 2.5380106132104794e-05,
82
+ "loss": 0.6373,
83
+ "step": 40
84
+ },
85
+ {
86
+ "epoch": 2.81,
87
+ "grad_norm": 0.9699375033378601,
88
+ "learning_rate": 2.4322601709933762e-05,
89
+ "loss": 0.5701,
90
+ "step": 45
91
+ },
92
+ {
93
+ "epoch": 3.0,
94
+ "eval_f1": 0.7326732673267327,
95
+ "eval_loss": 0.6987495422363281,
96
+ "eval_runtime": 1.344,
97
+ "eval_samples_per_second": 47.619,
98
+ "eval_steps_per_second": 5.952,
99
+ "step": 48
100
+ },
101
+ {
102
+ "epoch": 3.12,
103
+ "grad_norm": 0.5822675824165344,
104
+ "learning_rate": 2.3265097287762727e-05,
105
+ "loss": 0.5492,
106
+ "step": 50
107
+ },
108
+ {
109
+ "epoch": 3.44,
110
+ "grad_norm": 0.9859365224838257,
111
+ "learning_rate": 2.2207592865591692e-05,
112
+ "loss": 0.5552,
113
+ "step": 55
114
+ },
115
+ {
116
+ "epoch": 3.75,
117
+ "grad_norm": 0.7946156859397888,
118
+ "learning_rate": 2.115008844342066e-05,
119
+ "loss": 0.5459,
120
+ "step": 60
121
+ },
122
+ {
123
+ "epoch": 4.0,
124
+ "eval_f1": 0.7326732673267327,
125
+ "eval_loss": 0.6806888580322266,
126
+ "eval_runtime": 1.3463,
127
+ "eval_samples_per_second": 47.538,
128
+ "eval_steps_per_second": 5.942,
129
+ "step": 64
130
+ },
131
+ {
132
+ "epoch": 4.06,
133
+ "grad_norm": 1.2706314325332642,
134
+ "learning_rate": 2.009258402124963e-05,
135
+ "loss": 0.5792,
136
+ "step": 65
137
+ },
138
+ {
139
+ "epoch": 4.38,
140
+ "grad_norm": 1.456335186958313,
141
+ "learning_rate": 1.9035079599078593e-05,
142
+ "loss": 0.4967,
143
+ "step": 70
144
+ },
145
+ {
146
+ "epoch": 4.69,
147
+ "grad_norm": 1.6856915950775146,
148
+ "learning_rate": 1.797757517690756e-05,
149
+ "loss": 0.4966,
150
+ "step": 75
151
+ },
152
+ {
153
+ "epoch": 5.0,
154
+ "grad_norm": 4.080612659454346,
155
+ "learning_rate": 1.692007075473653e-05,
156
+ "loss": 0.4589,
157
+ "step": 80
158
+ },
159
+ {
160
+ "epoch": 5.0,
161
+ "eval_f1": 0.7474747474747475,
162
+ "eval_loss": 0.6495094299316406,
163
+ "eval_runtime": 1.348,
164
+ "eval_samples_per_second": 47.477,
165
+ "eval_steps_per_second": 5.935,
166
+ "step": 80
167
+ },
168
+ {
169
+ "epoch": 5.31,
170
+ "grad_norm": 2.8024661540985107,
171
+ "learning_rate": 1.5862566332565498e-05,
172
+ "loss": 0.4312,
173
+ "step": 85
174
+ },
175
+ {
176
+ "epoch": 5.62,
177
+ "grad_norm": 3.024564027786255,
178
+ "learning_rate": 1.4805061910394462e-05,
179
+ "loss": 0.4318,
180
+ "step": 90
181
+ },
182
+ {
183
+ "epoch": 5.94,
184
+ "grad_norm": 3.236055374145508,
185
+ "learning_rate": 1.3747557488223429e-05,
186
+ "loss": 0.3968,
187
+ "step": 95
188
+ },
189
+ {
190
+ "epoch": 6.0,
191
+ "eval_f1": 0.7391304347826088,
192
+ "eval_loss": 0.6395978927612305,
193
+ "eval_runtime": 1.3605,
194
+ "eval_samples_per_second": 47.043,
195
+ "eval_steps_per_second": 5.88,
196
+ "step": 96
197
+ },
198
+ {
199
+ "epoch": 6.25,
200
+ "grad_norm": 2.6648523807525635,
201
+ "learning_rate": 1.2690053066052397e-05,
202
+ "loss": 0.3623,
203
+ "step": 100
204
+ },
205
+ {
206
+ "epoch": 6.56,
207
+ "grad_norm": 2.6884357929229736,
208
+ "learning_rate": 1.1632548643881364e-05,
209
+ "loss": 0.4061,
210
+ "step": 105
211
+ },
212
+ {
213
+ "epoch": 6.88,
214
+ "grad_norm": 3.5860610008239746,
215
+ "learning_rate": 1.057504422171033e-05,
216
+ "loss": 0.3356,
217
+ "step": 110
218
+ },
219
+ {
220
+ "epoch": 7.0,
221
+ "eval_f1": 0.7659574468085106,
222
+ "eval_loss": 0.7118406295776367,
223
+ "eval_runtime": 1.3559,
224
+ "eval_samples_per_second": 47.202,
225
+ "eval_steps_per_second": 5.9,
226
+ "step": 112
227
+ },
228
+ {
229
+ "epoch": 7.19,
230
+ "grad_norm": 3.730743169784546,
231
+ "learning_rate": 9.517539799539297e-06,
232
+ "loss": 0.298,
233
+ "step": 115
234
+ },
235
+ {
236
+ "epoch": 7.5,
237
+ "grad_norm": 5.971029281616211,
238
+ "learning_rate": 8.460035377368265e-06,
239
+ "loss": 0.3183,
240
+ "step": 120
241
+ },
242
+ {
243
+ "epoch": 7.81,
244
+ "grad_norm": 3.3043053150177,
245
+ "learning_rate": 7.402530955197231e-06,
246
+ "loss": 0.2788,
247
+ "step": 125
248
+ },
249
+ {
250
+ "epoch": 8.0,
251
+ "eval_f1": 0.7272727272727273,
252
+ "eval_loss": 0.6860837936401367,
253
+ "eval_runtime": 1.3424,
254
+ "eval_samples_per_second": 47.676,
255
+ "eval_steps_per_second": 5.96,
256
+ "step": 128
257
+ },
258
+ {
259
+ "epoch": 8.12,
260
+ "grad_norm": 3.310666084289551,
261
+ "learning_rate": 6.556527417460405e-06,
262
+ "loss": 0.3052,
263
+ "step": 130
264
+ },
265
+ {
266
+ "epoch": 8.44,
267
+ "grad_norm": 3.9955413341522217,
268
+ "learning_rate": 5.499022995289372e-06,
269
+ "loss": 0.2833,
270
+ "step": 135
271
+ },
272
+ {
273
+ "epoch": 8.75,
274
+ "grad_norm": 2.4762678146362305,
275
+ "learning_rate": 4.441518573118339e-06,
276
+ "loss": 0.2387,
277
+ "step": 140
278
+ },
279
+ {
280
+ "epoch": 9.0,
281
+ "eval_f1": 0.7333333333333333,
282
+ "eval_loss": 0.7196941375732422,
283
+ "eval_runtime": 1.3575,
284
+ "eval_samples_per_second": 47.147,
285
+ "eval_steps_per_second": 5.893,
286
+ "step": 144
287
+ },
288
+ {
289
+ "epoch": 9.06,
290
+ "grad_norm": 5.2802581787109375,
291
+ "learning_rate": 3.3840141509473056e-06,
292
+ "loss": 0.3212,
293
+ "step": 145
294
+ },
295
+ {
296
+ "epoch": 9.38,
297
+ "grad_norm": 4.355774402618408,
298
+ "learning_rate": 2.326509728776273e-06,
299
+ "loss": 0.2311,
300
+ "step": 150
301
+ },
302
+ {
303
+ "epoch": 9.69,
304
+ "grad_norm": 3.7583141326904297,
305
+ "learning_rate": 1.2690053066052396e-06,
306
+ "loss": 0.2676,
307
+ "step": 155
308
+ },
309
+ {
310
+ "epoch": 10.0,
311
+ "grad_norm": 3.6632018089294434,
312
+ "learning_rate": 2.115008844342066e-07,
313
+ "loss": 0.254,
314
+ "step": 160
315
+ },
316
+ {
317
+ "epoch": 10.0,
318
+ "eval_f1": 0.7472527472527473,
319
+ "eval_loss": 0.7351865768432617,
320
+ "eval_runtime": 1.353,
321
+ "eval_samples_per_second": 47.304,
322
+ "eval_steps_per_second": 5.913,
323
+ "step": 160
324
+ }
325
+ ],
326
+ "logging_steps": 5,
327
+ "max_steps": 160,
328
+ "num_input_tokens_seen": 0,
329
+ "num_train_epochs": 10,
330
+ "save_steps": 500,
331
+ "total_flos": 8990959054911696.0,
332
+ "train_batch_size": 12,
333
+ "trial_name": null,
334
+ "trial_params": {
335
+ "learning_rate": 3.045612735852575e-05,
336
+ "per_device_train_batch_size": 12
337
+ }
338
+ }
run-2/checkpoint-160/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c88b324e8cf31191efb1727463a9a5c250d242da2e762cbea403fc8ef4c82b88
3
+ size 4920
run-2/checkpoint-32/config.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "ntu-spml/distilhubert",
3
+ "activation_dropout": 0.1,
4
+ "apply_spec_augment": false,
5
+ "architectures": [
6
+ "HubertForSequenceClassification"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 1,
10
+ "classifier_proj_size": 256,
11
+ "conv_bias": false,
12
+ "conv_dim": [
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512
20
+ ],
21
+ "conv_kernel": [
22
+ 10,
23
+ 3,
24
+ 3,
25
+ 3,
26
+ 3,
27
+ 2,
28
+ 2
29
+ ],
30
+ "conv_stride": [
31
+ 5,
32
+ 2,
33
+ 2,
34
+ 2,
35
+ 2,
36
+ 2,
37
+ 2
38
+ ],
39
+ "ctc_loss_reduction": "sum",
40
+ "ctc_zero_infinity": false,
41
+ "do_stable_layer_norm": false,
42
+ "eos_token_id": 2,
43
+ "feat_extract_activation": "gelu",
44
+ "feat_extract_norm": "group",
45
+ "feat_proj_dropout": 0.0,
46
+ "feat_proj_layer_norm": false,
47
+ "final_dropout": 0.0,
48
+ "hidden_act": "gelu",
49
+ "hidden_dropout": 0.1,
50
+ "hidden_size": 768,
51
+ "id2label": {
52
+ "0": "NOT_WORD",
53
+ "1": "WORD"
54
+ },
55
+ "initializer_range": 0.02,
56
+ "intermediate_size": 3072,
57
+ "label2id": {
58
+ "NOT_WORD": "0",
59
+ "WORD": "1"
60
+ },
61
+ "layer_norm_eps": 1e-05,
62
+ "layerdrop": 0.0,
63
+ "mask_feature_length": 10,
64
+ "mask_feature_min_masks": 0,
65
+ "mask_feature_prob": 0.0,
66
+ "mask_time_length": 10,
67
+ "mask_time_min_masks": 2,
68
+ "mask_time_prob": 0.05,
69
+ "model_type": "hubert",
70
+ "num_attention_heads": 12,
71
+ "num_conv_pos_embedding_groups": 16,
72
+ "num_conv_pos_embeddings": 128,
73
+ "num_feat_extract_layers": 7,
74
+ "num_hidden_layers": 2,
75
+ "pad_token_id": 0,
76
+ "torch_dtype": "float32",
77
+ "transformers_version": "4.38.1",
78
+ "use_weighted_layer_sum": false,
79
+ "vocab_size": 32
80
+ }
run-2/checkpoint-32/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e648e0a870999f989ef19ccc2c7c6d0f77ccc2010c289cd8bc12b80a7a066bca
3
+ size 94763496
run-2/checkpoint-32/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:002bb6c171e29c4e657fdf609d095f37e82aad84eb313cf2d9beae8f264b40ad
3
+ size 189552570
run-2/checkpoint-32/preprocessor_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0,
7
+ "return_attention_mask": true,
8
+ "sampling_rate": 16000
9
+ }
run-2/checkpoint-32/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:754d53121980eacb00801af5e57714a1cb2c0271b33c592667ed8ec1e79458ee
3
+ size 14244
run-2/checkpoint-32/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c4676fb94bec1071b8f1f8ce77bc6ff137b9ed2d1f7159edbb284bac270e1898
3
+ size 1064
run-2/checkpoint-32/trainer_state.json ADDED
@@ -0,0 +1,84 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.7326732673267327,
3
+ "best_model_checkpoint": "distilhubert-finetuned-not-a-word2/run-2/checkpoint-16",
4
+ "epoch": 2.0,
5
+ "eval_steps": 500,
6
+ "global_step": 32,
7
+ "is_hyper_param_search": true,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.31,
13
+ "grad_norm": 0.8302626013755798,
14
+ "learning_rate": 9.517539799539297e-06,
15
+ "loss": 0.6966,
16
+ "step": 5
17
+ },
18
+ {
19
+ "epoch": 0.62,
20
+ "grad_norm": 0.6829280257225037,
21
+ "learning_rate": 1.9035079599078593e-05,
22
+ "loss": 0.6813,
23
+ "step": 10
24
+ },
25
+ {
26
+ "epoch": 0.94,
27
+ "grad_norm": 0.5496665835380554,
28
+ "learning_rate": 2.8552619398617893e-05,
29
+ "loss": 0.6451,
30
+ "step": 15
31
+ },
32
+ {
33
+ "epoch": 1.0,
34
+ "eval_f1": 0.7326732673267327,
35
+ "eval_loss": 0.6860618591308594,
36
+ "eval_runtime": 1.3401,
37
+ "eval_samples_per_second": 47.759,
38
+ "eval_steps_per_second": 5.97,
39
+ "step": 16
40
+ },
41
+ {
42
+ "epoch": 1.25,
43
+ "grad_norm": 1.3569624423980713,
44
+ "learning_rate": 2.9610123820788925e-05,
45
+ "loss": 0.6113,
46
+ "step": 20
47
+ },
48
+ {
49
+ "epoch": 1.56,
50
+ "grad_norm": 0.3838456869125366,
51
+ "learning_rate": 2.8552619398617893e-05,
52
+ "loss": 0.5997,
53
+ "step": 25
54
+ },
55
+ {
56
+ "epoch": 1.88,
57
+ "grad_norm": 0.46842828392982483,
58
+ "learning_rate": 2.7495114976446858e-05,
59
+ "loss": 0.6133,
60
+ "step": 30
61
+ },
62
+ {
63
+ "epoch": 2.0,
64
+ "eval_f1": 0.7326732673267327,
65
+ "eval_loss": 0.719573974609375,
66
+ "eval_runtime": 1.3426,
67
+ "eval_samples_per_second": 47.668,
68
+ "eval_steps_per_second": 5.958,
69
+ "step": 32
70
+ }
71
+ ],
72
+ "logging_steps": 5,
73
+ "max_steps": 160,
74
+ "num_input_tokens_seen": 0,
75
+ "num_train_epochs": 10,
76
+ "save_steps": 500,
77
+ "total_flos": 1691273188513872.0,
78
+ "train_batch_size": 12,
79
+ "trial_name": null,
80
+ "trial_params": {
81
+ "learning_rate": 3.045612735852575e-05,
82
+ "per_device_train_batch_size": 12
83
+ }
84
+ }
run-2/checkpoint-32/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c88b324e8cf31191efb1727463a9a5c250d242da2e762cbea403fc8ef4c82b88
3
+ size 4920
run-2/checkpoint-48/config.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "ntu-spml/distilhubert",
3
+ "activation_dropout": 0.1,
4
+ "apply_spec_augment": false,
5
+ "architectures": [
6
+ "HubertForSequenceClassification"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 1,
10
+ "classifier_proj_size": 256,
11
+ "conv_bias": false,
12
+ "conv_dim": [
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512
20
+ ],
21
+ "conv_kernel": [
22
+ 10,
23
+ 3,
24
+ 3,
25
+ 3,
26
+ 3,
27
+ 2,
28
+ 2
29
+ ],
30
+ "conv_stride": [
31
+ 5,
32
+ 2,
33
+ 2,
34
+ 2,
35
+ 2,
36
+ 2,
37
+ 2
38
+ ],
39
+ "ctc_loss_reduction": "sum",
40
+ "ctc_zero_infinity": false,
41
+ "do_stable_layer_norm": false,
42
+ "eos_token_id": 2,
43
+ "feat_extract_activation": "gelu",
44
+ "feat_extract_norm": "group",
45
+ "feat_proj_dropout": 0.0,
46
+ "feat_proj_layer_norm": false,
47
+ "final_dropout": 0.0,
48
+ "hidden_act": "gelu",
49
+ "hidden_dropout": 0.1,
50
+ "hidden_size": 768,
51
+ "id2label": {
52
+ "0": "NOT_WORD",
53
+ "1": "WORD"
54
+ },
55
+ "initializer_range": 0.02,
56
+ "intermediate_size": 3072,
57
+ "label2id": {
58
+ "NOT_WORD": "0",
59
+ "WORD": "1"
60
+ },
61
+ "layer_norm_eps": 1e-05,
62
+ "layerdrop": 0.0,
63
+ "mask_feature_length": 10,
64
+ "mask_feature_min_masks": 0,
65
+ "mask_feature_prob": 0.0,
66
+ "mask_time_length": 10,
67
+ "mask_time_min_masks": 2,
68
+ "mask_time_prob": 0.05,
69
+ "model_type": "hubert",
70
+ "num_attention_heads": 12,
71
+ "num_conv_pos_embedding_groups": 16,
72
+ "num_conv_pos_embeddings": 128,
73
+ "num_feat_extract_layers": 7,
74
+ "num_hidden_layers": 2,
75
+ "pad_token_id": 0,
76
+ "torch_dtype": "float32",
77
+ "transformers_version": "4.38.1",
78
+ "use_weighted_layer_sum": false,
79
+ "vocab_size": 32
80
+ }
run-2/checkpoint-48/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5ac2283bb2c12ecd6c9c49788aaec50c130662cae463ed5a793ebb732f2e8ea
3
+ size 94763496
run-2/checkpoint-48/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cad2ee4ceb9e27e89f0b295c9158a445beb31acf5ce44b72af56615c9c00b33f
3
+ size 189552570
run-2/checkpoint-48/preprocessor_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0,
7
+ "return_attention_mask": true,
8
+ "sampling_rate": 16000
9
+ }
run-2/checkpoint-48/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6057f3e71568989f3d2442d841f7f161902200ee453a6d60795ac4142ad66214
3
+ size 14244
run-2/checkpoint-48/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f7b7c69a3d828af16e5c227fd151a2e419706d82611554e446439e8ee5183ce9
3
+ size 1064
run-2/checkpoint-48/trainer_state.json ADDED
@@ -0,0 +1,114 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.7326732673267327,
3
+ "best_model_checkpoint": "distilhubert-finetuned-not-a-word2/run-2/checkpoint-16",
4
+ "epoch": 3.0,
5
+ "eval_steps": 500,
6
+ "global_step": 48,
7
+ "is_hyper_param_search": true,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.31,
13
+ "grad_norm": 0.8302626013755798,
14
+ "learning_rate": 9.517539799539297e-06,
15
+ "loss": 0.6966,
16
+ "step": 5
17
+ },
18
+ {
19
+ "epoch": 0.62,
20
+ "grad_norm": 0.6829280257225037,
21
+ "learning_rate": 1.9035079599078593e-05,
22
+ "loss": 0.6813,
23
+ "step": 10
24
+ },
25
+ {
26
+ "epoch": 0.94,
27
+ "grad_norm": 0.5496665835380554,
28
+ "learning_rate": 2.8552619398617893e-05,
29
+ "loss": 0.6451,
30
+ "step": 15
31
+ },
32
+ {
33
+ "epoch": 1.0,
34
+ "eval_f1": 0.7326732673267327,
35
+ "eval_loss": 0.6860618591308594,
36
+ "eval_runtime": 1.3401,
37
+ "eval_samples_per_second": 47.759,
38
+ "eval_steps_per_second": 5.97,
39
+ "step": 16
40
+ },
41
+ {
42
+ "epoch": 1.25,
43
+ "grad_norm": 1.3569624423980713,
44
+ "learning_rate": 2.9610123820788925e-05,
45
+ "loss": 0.6113,
46
+ "step": 20
47
+ },
48
+ {
49
+ "epoch": 1.56,
50
+ "grad_norm": 0.3838456869125366,
51
+ "learning_rate": 2.8552619398617893e-05,
52
+ "loss": 0.5997,
53
+ "step": 25
54
+ },
55
+ {
56
+ "epoch": 1.88,
57
+ "grad_norm": 0.46842828392982483,
58
+ "learning_rate": 2.7495114976446858e-05,
59
+ "loss": 0.6133,
60
+ "step": 30
61
+ },
62
+ {
63
+ "epoch": 2.0,
64
+ "eval_f1": 0.7326732673267327,
65
+ "eval_loss": 0.719573974609375,
66
+ "eval_runtime": 1.3426,
67
+ "eval_samples_per_second": 47.668,
68
+ "eval_steps_per_second": 5.958,
69
+ "step": 32
70
+ },
71
+ {
72
+ "epoch": 2.19,
73
+ "grad_norm": 0.7610273957252502,
74
+ "learning_rate": 2.6437610554275826e-05,
75
+ "loss": 0.5971,
76
+ "step": 35
77
+ },
78
+ {
79
+ "epoch": 2.5,
80
+ "grad_norm": 0.5527757406234741,
81
+ "learning_rate": 2.5380106132104794e-05,
82
+ "loss": 0.6373,
83
+ "step": 40
84
+ },
85
+ {
86
+ "epoch": 2.81,
87
+ "grad_norm": 0.9699375033378601,
88
+ "learning_rate": 2.4322601709933762e-05,
89
+ "loss": 0.5701,
90
+ "step": 45
91
+ },
92
+ {
93
+ "epoch": 3.0,
94
+ "eval_f1": 0.7326732673267327,
95
+ "eval_loss": 0.6987495422363281,
96
+ "eval_runtime": 1.344,
97
+ "eval_samples_per_second": 47.619,
98
+ "eval_steps_per_second": 5.952,
99
+ "step": 48
100
+ }
101
+ ],
102
+ "logging_steps": 5,
103
+ "max_steps": 160,
104
+ "num_input_tokens_seen": 0,
105
+ "num_train_epochs": 10,
106
+ "save_steps": 500,
107
+ "total_flos": 2550470417237664.0,
108
+ "train_batch_size": 12,
109
+ "trial_name": null,
110
+ "trial_params": {
111
+ "learning_rate": 3.045612735852575e-05,
112
+ "per_device_train_batch_size": 12
113
+ }
114
+ }
run-2/checkpoint-48/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c88b324e8cf31191efb1727463a9a5c250d242da2e762cbea403fc8ef4c82b88
3
+ size 4920
run-2/checkpoint-64/config.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "ntu-spml/distilhubert",
3
+ "activation_dropout": 0.1,
4
+ "apply_spec_augment": false,
5
+ "architectures": [
6
+ "HubertForSequenceClassification"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 1,
10
+ "classifier_proj_size": 256,
11
+ "conv_bias": false,
12
+ "conv_dim": [
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512
20
+ ],
21
+ "conv_kernel": [
22
+ 10,
23
+ 3,
24
+ 3,
25
+ 3,
26
+ 3,
27
+ 2,
28
+ 2
29
+ ],
30
+ "conv_stride": [
31
+ 5,
32
+ 2,
33
+ 2,
34
+ 2,
35
+ 2,
36
+ 2,
37
+ 2
38
+ ],
39
+ "ctc_loss_reduction": "sum",
40
+ "ctc_zero_infinity": false,
41
+ "do_stable_layer_norm": false,
42
+ "eos_token_id": 2,
43
+ "feat_extract_activation": "gelu",
44
+ "feat_extract_norm": "group",
45
+ "feat_proj_dropout": 0.0,
46
+ "feat_proj_layer_norm": false,
47
+ "final_dropout": 0.0,
48
+ "hidden_act": "gelu",
49
+ "hidden_dropout": 0.1,
50
+ "hidden_size": 768,
51
+ "id2label": {
52
+ "0": "NOT_WORD",
53
+ "1": "WORD"
54
+ },
55
+ "initializer_range": 0.02,
56
+ "intermediate_size": 3072,
57
+ "label2id": {
58
+ "NOT_WORD": "0",
59
+ "WORD": "1"
60
+ },
61
+ "layer_norm_eps": 1e-05,
62
+ "layerdrop": 0.0,
63
+ "mask_feature_length": 10,
64
+ "mask_feature_min_masks": 0,
65
+ "mask_feature_prob": 0.0,
66
+ "mask_time_length": 10,
67
+ "mask_time_min_masks": 2,
68
+ "mask_time_prob": 0.05,
69
+ "model_type": "hubert",
70
+ "num_attention_heads": 12,
71
+ "num_conv_pos_embedding_groups": 16,
72
+ "num_conv_pos_embeddings": 128,
73
+ "num_feat_extract_layers": 7,
74
+ "num_hidden_layers": 2,
75
+ "pad_token_id": 0,
76
+ "torch_dtype": "float32",
77
+ "transformers_version": "4.38.1",
78
+ "use_weighted_layer_sum": false,
79
+ "vocab_size": 32
80
+ }