JaparSidik4 commited on
Commit
e9d3992
·
verified ·
1 Parent(s): ff0eb53

Upload folder using huggingface_hub

Browse files
Files changed (48) hide show
  1. checkpoint-1000/added_tokens.json +4 -0
  2. checkpoint-1000/config.json +92 -0
  3. checkpoint-1000/model.safetensors +3 -0
  4. checkpoint-1000/optimizer.pt +3 -0
  5. checkpoint-1000/rng_state.pth +3 -0
  6. checkpoint-1000/scaler.pt +3 -0
  7. checkpoint-1000/scheduler.pt +3 -0
  8. checkpoint-1000/special_tokens_map.json +13 -0
  9. checkpoint-1000/spm_char.model +3 -0
  10. checkpoint-1000/tokenizer_config.json +64 -0
  11. checkpoint-1000/trainer_state.json +321 -0
  12. checkpoint-1000/training_args.bin +3 -0
  13. checkpoint-2000/added_tokens.json +4 -0
  14. checkpoint-2000/config.json +92 -0
  15. checkpoint-2000/model.safetensors +3 -0
  16. checkpoint-2000/optimizer.pt +3 -0
  17. checkpoint-2000/rng_state.pth +3 -0
  18. checkpoint-2000/scaler.pt +3 -0
  19. checkpoint-2000/scheduler.pt +3 -0
  20. checkpoint-2000/special_tokens_map.json +13 -0
  21. checkpoint-2000/spm_char.model +3 -0
  22. checkpoint-2000/tokenizer_config.json +64 -0
  23. checkpoint-2000/trainer_state.json +609 -0
  24. checkpoint-2000/training_args.bin +3 -0
  25. checkpoint-3000/added_tokens.json +4 -0
  26. checkpoint-3000/config.json +92 -0
  27. checkpoint-3000/model.safetensors +3 -0
  28. checkpoint-3000/optimizer.pt +3 -0
  29. checkpoint-3000/rng_state.pth +3 -0
  30. checkpoint-3000/scaler.pt +3 -0
  31. checkpoint-3000/scheduler.pt +3 -0
  32. checkpoint-3000/special_tokens_map.json +13 -0
  33. checkpoint-3000/spm_char.model +3 -0
  34. checkpoint-3000/tokenizer_config.json +64 -0
  35. checkpoint-3000/trainer_state.json +897 -0
  36. checkpoint-3000/training_args.bin +3 -0
  37. checkpoint-4000/added_tokens.json +4 -0
  38. checkpoint-4000/config.json +92 -0
  39. checkpoint-4000/model.safetensors +3 -0
  40. checkpoint-4000/optimizer.pt +3 -0
  41. checkpoint-4000/rng_state.pth +3 -0
  42. checkpoint-4000/scaler.pt +3 -0
  43. checkpoint-4000/scheduler.pt +3 -0
  44. checkpoint-4000/special_tokens_map.json +13 -0
  45. checkpoint-4000/spm_char.model +3 -0
  46. checkpoint-4000/tokenizer_config.json +64 -0
  47. checkpoint-4000/trainer_state.json +1185 -0
  48. checkpoint-4000/training_args.bin +3 -0
checkpoint-1000/added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "<ctc_blank>": 80,
3
+ "<mask>": 79
4
+ }
checkpoint-1000/config.json ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/speecht5_tts",
3
+ "activation_dropout": 0.1,
4
+ "apply_spec_augment": true,
5
+ "architectures": [
6
+ "SpeechT5ForTextToSpeech"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 0,
10
+ "conv_bias": false,
11
+ "conv_dim": [
12
+ 512,
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512
19
+ ],
20
+ "conv_kernel": [
21
+ 10,
22
+ 3,
23
+ 3,
24
+ 3,
25
+ 3,
26
+ 2,
27
+ 2
28
+ ],
29
+ "conv_stride": [
30
+ 5,
31
+ 2,
32
+ 2,
33
+ 2,
34
+ 2,
35
+ 2,
36
+ 2
37
+ ],
38
+ "decoder_attention_heads": 12,
39
+ "decoder_ffn_dim": 3072,
40
+ "decoder_layerdrop": 0.1,
41
+ "decoder_layers": 6,
42
+ "decoder_start_token_id": 2,
43
+ "encoder_attention_heads": 12,
44
+ "encoder_ffn_dim": 3072,
45
+ "encoder_layerdrop": 0.1,
46
+ "encoder_layers": 12,
47
+ "encoder_max_relative_position": 160,
48
+ "eos_token_id": 2,
49
+ "feat_extract_activation": "gelu",
50
+ "feat_extract_norm": "group",
51
+ "feat_proj_dropout": 0.0,
52
+ "guided_attention_loss_num_heads": 2,
53
+ "guided_attention_loss_scale": 10.0,
54
+ "guided_attention_loss_sigma": 0.4,
55
+ "hidden_act": "gelu",
56
+ "hidden_dropout": 0.1,
57
+ "hidden_size": 768,
58
+ "initializer_range": 0.02,
59
+ "is_encoder_decoder": true,
60
+ "layer_norm_eps": 1e-05,
61
+ "mask_feature_length": 10,
62
+ "mask_feature_min_masks": 0,
63
+ "mask_feature_prob": 0.0,
64
+ "mask_time_length": 10,
65
+ "mask_time_min_masks": 2,
66
+ "mask_time_prob": 0.05,
67
+ "max_length": 1876,
68
+ "max_speech_positions": 1876,
69
+ "max_text_positions": 600,
70
+ "model_type": "speecht5",
71
+ "num_conv_pos_embedding_groups": 16,
72
+ "num_conv_pos_embeddings": 128,
73
+ "num_feat_extract_layers": 7,
74
+ "num_mel_bins": 80,
75
+ "pad_token_id": 1,
76
+ "positional_dropout": 0.1,
77
+ "reduction_factor": 2,
78
+ "scale_embedding": false,
79
+ "speaker_embedding_dim": 512,
80
+ "speech_decoder_postnet_dropout": 0.5,
81
+ "speech_decoder_postnet_kernel": 5,
82
+ "speech_decoder_postnet_layers": 5,
83
+ "speech_decoder_postnet_units": 256,
84
+ "speech_decoder_prenet_dropout": 0.5,
85
+ "speech_decoder_prenet_layers": 2,
86
+ "speech_decoder_prenet_units": 256,
87
+ "torch_dtype": "float32",
88
+ "transformers_version": "4.50.0.dev0",
89
+ "use_cache": false,
90
+ "use_guided_attention_loss": true,
91
+ "vocab_size": 81
92
+ }
checkpoint-1000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33cbbe0323c17fa0dd71cd2be9b960ae209a74e0c8e874a3d66f1b1921059f37
3
+ size 577789320
checkpoint-1000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f1314809864351fb28e22d1c4f0ac2574355a1164728336ff9f7690f48e444e7
3
+ size 1155772233
checkpoint-1000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:585ca82d8e14932ad03539b98b5915a62b45c3c2faf89e0a54708e90b136e1d0
3
+ size 14244
checkpoint-1000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87aa44525984371790d54d96d799fdac8e4a1fd53749608cf9dcf03ca2c78583
3
+ size 988
checkpoint-1000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:04adfa1aca9f7b9ed4b807866282ff1495e43e68a72612345d80bdb13c2a1305
3
+ size 1064
checkpoint-1000/special_tokens_map.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "eos_token": "</s>",
4
+ "mask_token": {
5
+ "content": "<mask>",
6
+ "lstrip": true,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false
10
+ },
11
+ "pad_token": "<pad>",
12
+ "unk_token": "<unk>"
13
+ }
checkpoint-1000/spm_char.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7fcc48f3e225f627b1641db410ceb0c8649bd2b0c982e150b03f8be3728ab560
3
+ size 238473
checkpoint-1000/tokenizer_config.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "79": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": true,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "80": {
44
+ "content": "<ctc_blank>",
45
+ "lstrip": false,
46
+ "normalized": true,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": false
50
+ }
51
+ },
52
+ "bos_token": "<s>",
53
+ "clean_up_tokenization_spaces": false,
54
+ "eos_token": "</s>",
55
+ "extra_special_tokens": {},
56
+ "mask_token": "<mask>",
57
+ "model_max_length": 600,
58
+ "normalize": false,
59
+ "pad_token": "<pad>",
60
+ "processor_class": "SpeechT5Processor",
61
+ "sp_model_kwargs": {},
62
+ "tokenizer_class": "SpeechT5Tokenizer",
63
+ "unk_token": "<unk>"
64
+ }
checkpoint-1000/trainer_state.json ADDED
@@ -0,0 +1,321 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.4698790907859802,
3
+ "best_model_checkpoint": "./speecht5_tts_common_voice_17_0_id/checkpoint-1000",
4
+ "epoch": 7.142857142857143,
5
+ "eval_steps": 1000,
6
+ "global_step": 1000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.17857142857142858,
13
+ "grad_norm": 4.796061038970947,
14
+ "learning_rate": 4.6000000000000004e-07,
15
+ "loss": 1.0232,
16
+ "step": 25
17
+ },
18
+ {
19
+ "epoch": 0.35714285714285715,
20
+ "grad_norm": 6.88640022277832,
21
+ "learning_rate": 9.600000000000001e-07,
22
+ "loss": 0.9412,
23
+ "step": 50
24
+ },
25
+ {
26
+ "epoch": 0.5357142857142857,
27
+ "grad_norm": 4.282207489013672,
28
+ "learning_rate": 1.46e-06,
29
+ "loss": 0.8431,
30
+ "step": 75
31
+ },
32
+ {
33
+ "epoch": 0.7142857142857143,
34
+ "grad_norm": 4.095445156097412,
35
+ "learning_rate": 1.9600000000000003e-06,
36
+ "loss": 0.7959,
37
+ "step": 100
38
+ },
39
+ {
40
+ "epoch": 0.8928571428571429,
41
+ "grad_norm": 6.723930835723877,
42
+ "learning_rate": 2.46e-06,
43
+ "loss": 0.7825,
44
+ "step": 125
45
+ },
46
+ {
47
+ "epoch": 1.0714285714285714,
48
+ "grad_norm": 2.318833589553833,
49
+ "learning_rate": 2.96e-06,
50
+ "loss": 0.7967,
51
+ "step": 150
52
+ },
53
+ {
54
+ "epoch": 1.25,
55
+ "grad_norm": 2.1883444786071777,
56
+ "learning_rate": 3.46e-06,
57
+ "loss": 0.7826,
58
+ "step": 175
59
+ },
60
+ {
61
+ "epoch": 1.4285714285714286,
62
+ "grad_norm": 2.7793357372283936,
63
+ "learning_rate": 3.96e-06,
64
+ "loss": 0.7238,
65
+ "step": 200
66
+ },
67
+ {
68
+ "epoch": 1.6071428571428572,
69
+ "grad_norm": 3.0292282104492188,
70
+ "learning_rate": 4.4600000000000005e-06,
71
+ "loss": 0.6981,
72
+ "step": 225
73
+ },
74
+ {
75
+ "epoch": 1.7857142857142856,
76
+ "grad_norm": 4.8227314949035645,
77
+ "learning_rate": 4.960000000000001e-06,
78
+ "loss": 0.6753,
79
+ "step": 250
80
+ },
81
+ {
82
+ "epoch": 1.9642857142857144,
83
+ "grad_norm": 2.781235694885254,
84
+ "learning_rate": 5.460000000000001e-06,
85
+ "loss": 0.6337,
86
+ "step": 275
87
+ },
88
+ {
89
+ "epoch": 2.142857142857143,
90
+ "grad_norm": 2.9240171909332275,
91
+ "learning_rate": 5.9600000000000005e-06,
92
+ "loss": 0.6499,
93
+ "step": 300
94
+ },
95
+ {
96
+ "epoch": 2.3214285714285716,
97
+ "grad_norm": 2.0720555782318115,
98
+ "learning_rate": 6.460000000000001e-06,
99
+ "loss": 0.5995,
100
+ "step": 325
101
+ },
102
+ {
103
+ "epoch": 2.5,
104
+ "grad_norm": 3.025918483734131,
105
+ "learning_rate": 6.96e-06,
106
+ "loss": 0.6054,
107
+ "step": 350
108
+ },
109
+ {
110
+ "epoch": 2.678571428571429,
111
+ "grad_norm": 3.3047034740448,
112
+ "learning_rate": 7.4600000000000006e-06,
113
+ "loss": 0.5828,
114
+ "step": 375
115
+ },
116
+ {
117
+ "epoch": 2.857142857142857,
118
+ "grad_norm": 2.4087064266204834,
119
+ "learning_rate": 7.960000000000002e-06,
120
+ "loss": 0.5706,
121
+ "step": 400
122
+ },
123
+ {
124
+ "epoch": 3.0357142857142856,
125
+ "grad_norm": 2.7837984561920166,
126
+ "learning_rate": 8.46e-06,
127
+ "loss": 0.579,
128
+ "step": 425
129
+ },
130
+ {
131
+ "epoch": 3.2142857142857144,
132
+ "grad_norm": 1.7530593872070312,
133
+ "learning_rate": 8.96e-06,
134
+ "loss": 0.5455,
135
+ "step": 450
136
+ },
137
+ {
138
+ "epoch": 3.392857142857143,
139
+ "grad_norm": 2.3677515983581543,
140
+ "learning_rate": 9.460000000000001e-06,
141
+ "loss": 0.5557,
142
+ "step": 475
143
+ },
144
+ {
145
+ "epoch": 3.571428571428571,
146
+ "grad_norm": 2.445343494415283,
147
+ "learning_rate": 9.960000000000001e-06,
148
+ "loss": 0.5686,
149
+ "step": 500
150
+ },
151
+ {
152
+ "epoch": 3.75,
153
+ "grad_norm": 4.732877731323242,
154
+ "learning_rate": 9.934285714285715e-06,
155
+ "loss": 0.5635,
156
+ "step": 525
157
+ },
158
+ {
159
+ "epoch": 3.928571428571429,
160
+ "grad_norm": 4.45880126953125,
161
+ "learning_rate": 9.862857142857144e-06,
162
+ "loss": 0.5477,
163
+ "step": 550
164
+ },
165
+ {
166
+ "epoch": 4.107142857142857,
167
+ "grad_norm": 3.9657559394836426,
168
+ "learning_rate": 9.791428571428571e-06,
169
+ "loss": 0.5563,
170
+ "step": 575
171
+ },
172
+ {
173
+ "epoch": 4.285714285714286,
174
+ "grad_norm": 2.676309823989868,
175
+ "learning_rate": 9.72e-06,
176
+ "loss": 0.5221,
177
+ "step": 600
178
+ },
179
+ {
180
+ "epoch": 4.464285714285714,
181
+ "grad_norm": 2.89937686920166,
182
+ "learning_rate": 9.648571428571429e-06,
183
+ "loss": 0.5211,
184
+ "step": 625
185
+ },
186
+ {
187
+ "epoch": 4.642857142857143,
188
+ "grad_norm": 2.9824204444885254,
189
+ "learning_rate": 9.577142857142858e-06,
190
+ "loss": 0.5583,
191
+ "step": 650
192
+ },
193
+ {
194
+ "epoch": 4.821428571428571,
195
+ "grad_norm": 3.632369041442871,
196
+ "learning_rate": 9.505714285714287e-06,
197
+ "loss": 0.526,
198
+ "step": 675
199
+ },
200
+ {
201
+ "epoch": 5.0,
202
+ "grad_norm": 2.352837324142456,
203
+ "learning_rate": 9.434285714285714e-06,
204
+ "loss": 0.5211,
205
+ "step": 700
206
+ },
207
+ {
208
+ "epoch": 5.178571428571429,
209
+ "grad_norm": 1.3236348628997803,
210
+ "learning_rate": 9.362857142857143e-06,
211
+ "loss": 0.5173,
212
+ "step": 725
213
+ },
214
+ {
215
+ "epoch": 5.357142857142857,
216
+ "grad_norm": 1.8493444919586182,
217
+ "learning_rate": 9.291428571428572e-06,
218
+ "loss": 0.5094,
219
+ "step": 750
220
+ },
221
+ {
222
+ "epoch": 5.535714285714286,
223
+ "grad_norm": 1.7403544187545776,
224
+ "learning_rate": 9.220000000000002e-06,
225
+ "loss": 0.5239,
226
+ "step": 775
227
+ },
228
+ {
229
+ "epoch": 5.714285714285714,
230
+ "grad_norm": 4.10603666305542,
231
+ "learning_rate": 9.148571428571429e-06,
232
+ "loss": 0.5304,
233
+ "step": 800
234
+ },
235
+ {
236
+ "epoch": 5.892857142857143,
237
+ "grad_norm": 2.69100284576416,
238
+ "learning_rate": 9.077142857142858e-06,
239
+ "loss": 0.5224,
240
+ "step": 825
241
+ },
242
+ {
243
+ "epoch": 6.071428571428571,
244
+ "grad_norm": 1.956499695777893,
245
+ "learning_rate": 9.005714285714287e-06,
246
+ "loss": 0.5065,
247
+ "step": 850
248
+ },
249
+ {
250
+ "epoch": 6.25,
251
+ "grad_norm": 1.915540337562561,
252
+ "learning_rate": 8.934285714285716e-06,
253
+ "loss": 0.511,
254
+ "step": 875
255
+ },
256
+ {
257
+ "epoch": 6.428571428571429,
258
+ "grad_norm": 2.1903812885284424,
259
+ "learning_rate": 8.862857142857143e-06,
260
+ "loss": 0.5132,
261
+ "step": 900
262
+ },
263
+ {
264
+ "epoch": 6.607142857142857,
265
+ "grad_norm": 1.666169285774231,
266
+ "learning_rate": 8.791428571428572e-06,
267
+ "loss": 0.5362,
268
+ "step": 925
269
+ },
270
+ {
271
+ "epoch": 6.785714285714286,
272
+ "grad_norm": 1.6485190391540527,
273
+ "learning_rate": 8.720000000000001e-06,
274
+ "loss": 0.5115,
275
+ "step": 950
276
+ },
277
+ {
278
+ "epoch": 6.964285714285714,
279
+ "grad_norm": 1.7726221084594727,
280
+ "learning_rate": 8.64857142857143e-06,
281
+ "loss": 0.5183,
282
+ "step": 975
283
+ },
284
+ {
285
+ "epoch": 7.142857142857143,
286
+ "grad_norm": 2.7793469429016113,
287
+ "learning_rate": 8.577142857142858e-06,
288
+ "loss": 0.5133,
289
+ "step": 1000
290
+ },
291
+ {
292
+ "epoch": 7.142857142857143,
293
+ "eval_loss": 0.4698790907859802,
294
+ "eval_runtime": 10.706,
295
+ "eval_samples_per_second": 46.422,
296
+ "eval_steps_per_second": 5.885,
297
+ "step": 1000
298
+ }
299
+ ],
300
+ "logging_steps": 25,
301
+ "max_steps": 4000,
302
+ "num_input_tokens_seen": 0,
303
+ "num_train_epochs": 29,
304
+ "save_steps": 1000,
305
+ "stateful_callbacks": {
306
+ "TrainerControl": {
307
+ "args": {
308
+ "should_epoch_stop": false,
309
+ "should_evaluate": false,
310
+ "should_log": false,
311
+ "should_save": true,
312
+ "should_training_stop": false
313
+ },
314
+ "attributes": {}
315
+ }
316
+ },
317
+ "total_flos": 2963090237724936.0,
318
+ "train_batch_size": 16,
319
+ "trial_name": null,
320
+ "trial_params": null
321
+ }
checkpoint-1000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a01e8c0a090b02699d5b506a30ffaf176234c63870a5a741f73db593b385337
3
+ size 5560
checkpoint-2000/added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "<ctc_blank>": 80,
3
+ "<mask>": 79
4
+ }
checkpoint-2000/config.json ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/speecht5_tts",
3
+ "activation_dropout": 0.1,
4
+ "apply_spec_augment": true,
5
+ "architectures": [
6
+ "SpeechT5ForTextToSpeech"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 0,
10
+ "conv_bias": false,
11
+ "conv_dim": [
12
+ 512,
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512
19
+ ],
20
+ "conv_kernel": [
21
+ 10,
22
+ 3,
23
+ 3,
24
+ 3,
25
+ 3,
26
+ 2,
27
+ 2
28
+ ],
29
+ "conv_stride": [
30
+ 5,
31
+ 2,
32
+ 2,
33
+ 2,
34
+ 2,
35
+ 2,
36
+ 2
37
+ ],
38
+ "decoder_attention_heads": 12,
39
+ "decoder_ffn_dim": 3072,
40
+ "decoder_layerdrop": 0.1,
41
+ "decoder_layers": 6,
42
+ "decoder_start_token_id": 2,
43
+ "encoder_attention_heads": 12,
44
+ "encoder_ffn_dim": 3072,
45
+ "encoder_layerdrop": 0.1,
46
+ "encoder_layers": 12,
47
+ "encoder_max_relative_position": 160,
48
+ "eos_token_id": 2,
49
+ "feat_extract_activation": "gelu",
50
+ "feat_extract_norm": "group",
51
+ "feat_proj_dropout": 0.0,
52
+ "guided_attention_loss_num_heads": 2,
53
+ "guided_attention_loss_scale": 10.0,
54
+ "guided_attention_loss_sigma": 0.4,
55
+ "hidden_act": "gelu",
56
+ "hidden_dropout": 0.1,
57
+ "hidden_size": 768,
58
+ "initializer_range": 0.02,
59
+ "is_encoder_decoder": true,
60
+ "layer_norm_eps": 1e-05,
61
+ "mask_feature_length": 10,
62
+ "mask_feature_min_masks": 0,
63
+ "mask_feature_prob": 0.0,
64
+ "mask_time_length": 10,
65
+ "mask_time_min_masks": 2,
66
+ "mask_time_prob": 0.05,
67
+ "max_length": 1876,
68
+ "max_speech_positions": 1876,
69
+ "max_text_positions": 600,
70
+ "model_type": "speecht5",
71
+ "num_conv_pos_embedding_groups": 16,
72
+ "num_conv_pos_embeddings": 128,
73
+ "num_feat_extract_layers": 7,
74
+ "num_mel_bins": 80,
75
+ "pad_token_id": 1,
76
+ "positional_dropout": 0.1,
77
+ "reduction_factor": 2,
78
+ "scale_embedding": false,
79
+ "speaker_embedding_dim": 512,
80
+ "speech_decoder_postnet_dropout": 0.5,
81
+ "speech_decoder_postnet_kernel": 5,
82
+ "speech_decoder_postnet_layers": 5,
83
+ "speech_decoder_postnet_units": 256,
84
+ "speech_decoder_prenet_dropout": 0.5,
85
+ "speech_decoder_prenet_layers": 2,
86
+ "speech_decoder_prenet_units": 256,
87
+ "torch_dtype": "float32",
88
+ "transformers_version": "4.50.0.dev0",
89
+ "use_cache": false,
90
+ "use_guided_attention_loss": true,
91
+ "vocab_size": 81
92
+ }
checkpoint-2000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8af5f93c52f7348e0dd980f2d568d4d1a57e81e501bfb716dda04491379bf257
3
+ size 577789320
checkpoint-2000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8db569bdc5f32ee3acb80da044109ad11a26cf8f0933e4978089b6ac2e34be34
3
+ size 1155772233
checkpoint-2000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ada26402e6f6f7923f2e77df566e5aa603082f9c0ae08d4066c59fd29adf6c17
3
+ size 14244
checkpoint-2000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c0b23ed8ad30394561c40ec7ebe1b25b72685906a15189a8b49897599d008289
3
+ size 988
checkpoint-2000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9892f0087eb652598b43152ddd8519208d4c860334ace0372ec523095d6aebd8
3
+ size 1064
checkpoint-2000/special_tokens_map.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "eos_token": "</s>",
4
+ "mask_token": {
5
+ "content": "<mask>",
6
+ "lstrip": true,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false
10
+ },
11
+ "pad_token": "<pad>",
12
+ "unk_token": "<unk>"
13
+ }
checkpoint-2000/spm_char.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7fcc48f3e225f627b1641db410ceb0c8649bd2b0c982e150b03f8be3728ab560
3
+ size 238473
checkpoint-2000/tokenizer_config.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "79": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": true,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "80": {
44
+ "content": "<ctc_blank>",
45
+ "lstrip": false,
46
+ "normalized": true,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": false
50
+ }
51
+ },
52
+ "bos_token": "<s>",
53
+ "clean_up_tokenization_spaces": false,
54
+ "eos_token": "</s>",
55
+ "extra_special_tokens": {},
56
+ "mask_token": "<mask>",
57
+ "model_max_length": 600,
58
+ "normalize": false,
59
+ "pad_token": "<pad>",
60
+ "processor_class": "SpeechT5Processor",
61
+ "sp_model_kwargs": {},
62
+ "tokenizer_class": "SpeechT5Tokenizer",
63
+ "unk_token": "<unk>"
64
+ }
checkpoint-2000/trainer_state.json ADDED
@@ -0,0 +1,609 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.45390585064888,
3
+ "best_model_checkpoint": "./speecht5_tts_common_voice_17_0_id/checkpoint-2000",
4
+ "epoch": 14.285714285714286,
5
+ "eval_steps": 1000,
6
+ "global_step": 2000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.17857142857142858,
13
+ "grad_norm": 4.796061038970947,
14
+ "learning_rate": 4.6000000000000004e-07,
15
+ "loss": 1.0232,
16
+ "step": 25
17
+ },
18
+ {
19
+ "epoch": 0.35714285714285715,
20
+ "grad_norm": 6.88640022277832,
21
+ "learning_rate": 9.600000000000001e-07,
22
+ "loss": 0.9412,
23
+ "step": 50
24
+ },
25
+ {
26
+ "epoch": 0.5357142857142857,
27
+ "grad_norm": 4.282207489013672,
28
+ "learning_rate": 1.46e-06,
29
+ "loss": 0.8431,
30
+ "step": 75
31
+ },
32
+ {
33
+ "epoch": 0.7142857142857143,
34
+ "grad_norm": 4.095445156097412,
35
+ "learning_rate": 1.9600000000000003e-06,
36
+ "loss": 0.7959,
37
+ "step": 100
38
+ },
39
+ {
40
+ "epoch": 0.8928571428571429,
41
+ "grad_norm": 6.723930835723877,
42
+ "learning_rate": 2.46e-06,
43
+ "loss": 0.7825,
44
+ "step": 125
45
+ },
46
+ {
47
+ "epoch": 1.0714285714285714,
48
+ "grad_norm": 2.318833589553833,
49
+ "learning_rate": 2.96e-06,
50
+ "loss": 0.7967,
51
+ "step": 150
52
+ },
53
+ {
54
+ "epoch": 1.25,
55
+ "grad_norm": 2.1883444786071777,
56
+ "learning_rate": 3.46e-06,
57
+ "loss": 0.7826,
58
+ "step": 175
59
+ },
60
+ {
61
+ "epoch": 1.4285714285714286,
62
+ "grad_norm": 2.7793357372283936,
63
+ "learning_rate": 3.96e-06,
64
+ "loss": 0.7238,
65
+ "step": 200
66
+ },
67
+ {
68
+ "epoch": 1.6071428571428572,
69
+ "grad_norm": 3.0292282104492188,
70
+ "learning_rate": 4.4600000000000005e-06,
71
+ "loss": 0.6981,
72
+ "step": 225
73
+ },
74
+ {
75
+ "epoch": 1.7857142857142856,
76
+ "grad_norm": 4.8227314949035645,
77
+ "learning_rate": 4.960000000000001e-06,
78
+ "loss": 0.6753,
79
+ "step": 250
80
+ },
81
+ {
82
+ "epoch": 1.9642857142857144,
83
+ "grad_norm": 2.781235694885254,
84
+ "learning_rate": 5.460000000000001e-06,
85
+ "loss": 0.6337,
86
+ "step": 275
87
+ },
88
+ {
89
+ "epoch": 2.142857142857143,
90
+ "grad_norm": 2.9240171909332275,
91
+ "learning_rate": 5.9600000000000005e-06,
92
+ "loss": 0.6499,
93
+ "step": 300
94
+ },
95
+ {
96
+ "epoch": 2.3214285714285716,
97
+ "grad_norm": 2.0720555782318115,
98
+ "learning_rate": 6.460000000000001e-06,
99
+ "loss": 0.5995,
100
+ "step": 325
101
+ },
102
+ {
103
+ "epoch": 2.5,
104
+ "grad_norm": 3.025918483734131,
105
+ "learning_rate": 6.96e-06,
106
+ "loss": 0.6054,
107
+ "step": 350
108
+ },
109
+ {
110
+ "epoch": 2.678571428571429,
111
+ "grad_norm": 3.3047034740448,
112
+ "learning_rate": 7.4600000000000006e-06,
113
+ "loss": 0.5828,
114
+ "step": 375
115
+ },
116
+ {
117
+ "epoch": 2.857142857142857,
118
+ "grad_norm": 2.4087064266204834,
119
+ "learning_rate": 7.960000000000002e-06,
120
+ "loss": 0.5706,
121
+ "step": 400
122
+ },
123
+ {
124
+ "epoch": 3.0357142857142856,
125
+ "grad_norm": 2.7837984561920166,
126
+ "learning_rate": 8.46e-06,
127
+ "loss": 0.579,
128
+ "step": 425
129
+ },
130
+ {
131
+ "epoch": 3.2142857142857144,
132
+ "grad_norm": 1.7530593872070312,
133
+ "learning_rate": 8.96e-06,
134
+ "loss": 0.5455,
135
+ "step": 450
136
+ },
137
+ {
138
+ "epoch": 3.392857142857143,
139
+ "grad_norm": 2.3677515983581543,
140
+ "learning_rate": 9.460000000000001e-06,
141
+ "loss": 0.5557,
142
+ "step": 475
143
+ },
144
+ {
145
+ "epoch": 3.571428571428571,
146
+ "grad_norm": 2.445343494415283,
147
+ "learning_rate": 9.960000000000001e-06,
148
+ "loss": 0.5686,
149
+ "step": 500
150
+ },
151
+ {
152
+ "epoch": 3.75,
153
+ "grad_norm": 4.732877731323242,
154
+ "learning_rate": 9.934285714285715e-06,
155
+ "loss": 0.5635,
156
+ "step": 525
157
+ },
158
+ {
159
+ "epoch": 3.928571428571429,
160
+ "grad_norm": 4.45880126953125,
161
+ "learning_rate": 9.862857142857144e-06,
162
+ "loss": 0.5477,
163
+ "step": 550
164
+ },
165
+ {
166
+ "epoch": 4.107142857142857,
167
+ "grad_norm": 3.9657559394836426,
168
+ "learning_rate": 9.791428571428571e-06,
169
+ "loss": 0.5563,
170
+ "step": 575
171
+ },
172
+ {
173
+ "epoch": 4.285714285714286,
174
+ "grad_norm": 2.676309823989868,
175
+ "learning_rate": 9.72e-06,
176
+ "loss": 0.5221,
177
+ "step": 600
178
+ },
179
+ {
180
+ "epoch": 4.464285714285714,
181
+ "grad_norm": 2.89937686920166,
182
+ "learning_rate": 9.648571428571429e-06,
183
+ "loss": 0.5211,
184
+ "step": 625
185
+ },
186
+ {
187
+ "epoch": 4.642857142857143,
188
+ "grad_norm": 2.9824204444885254,
189
+ "learning_rate": 9.577142857142858e-06,
190
+ "loss": 0.5583,
191
+ "step": 650
192
+ },
193
+ {
194
+ "epoch": 4.821428571428571,
195
+ "grad_norm": 3.632369041442871,
196
+ "learning_rate": 9.505714285714287e-06,
197
+ "loss": 0.526,
198
+ "step": 675
199
+ },
200
+ {
201
+ "epoch": 5.0,
202
+ "grad_norm": 2.352837324142456,
203
+ "learning_rate": 9.434285714285714e-06,
204
+ "loss": 0.5211,
205
+ "step": 700
206
+ },
207
+ {
208
+ "epoch": 5.178571428571429,
209
+ "grad_norm": 1.3236348628997803,
210
+ "learning_rate": 9.362857142857143e-06,
211
+ "loss": 0.5173,
212
+ "step": 725
213
+ },
214
+ {
215
+ "epoch": 5.357142857142857,
216
+ "grad_norm": 1.8493444919586182,
217
+ "learning_rate": 9.291428571428572e-06,
218
+ "loss": 0.5094,
219
+ "step": 750
220
+ },
221
+ {
222
+ "epoch": 5.535714285714286,
223
+ "grad_norm": 1.7403544187545776,
224
+ "learning_rate": 9.220000000000002e-06,
225
+ "loss": 0.5239,
226
+ "step": 775
227
+ },
228
+ {
229
+ "epoch": 5.714285714285714,
230
+ "grad_norm": 4.10603666305542,
231
+ "learning_rate": 9.148571428571429e-06,
232
+ "loss": 0.5304,
233
+ "step": 800
234
+ },
235
+ {
236
+ "epoch": 5.892857142857143,
237
+ "grad_norm": 2.69100284576416,
238
+ "learning_rate": 9.077142857142858e-06,
239
+ "loss": 0.5224,
240
+ "step": 825
241
+ },
242
+ {
243
+ "epoch": 6.071428571428571,
244
+ "grad_norm": 1.956499695777893,
245
+ "learning_rate": 9.005714285714287e-06,
246
+ "loss": 0.5065,
247
+ "step": 850
248
+ },
249
+ {
250
+ "epoch": 6.25,
251
+ "grad_norm": 1.915540337562561,
252
+ "learning_rate": 8.934285714285716e-06,
253
+ "loss": 0.511,
254
+ "step": 875
255
+ },
256
+ {
257
+ "epoch": 6.428571428571429,
258
+ "grad_norm": 2.1903812885284424,
259
+ "learning_rate": 8.862857142857143e-06,
260
+ "loss": 0.5132,
261
+ "step": 900
262
+ },
263
+ {
264
+ "epoch": 6.607142857142857,
265
+ "grad_norm": 1.666169285774231,
266
+ "learning_rate": 8.791428571428572e-06,
267
+ "loss": 0.5362,
268
+ "step": 925
269
+ },
270
+ {
271
+ "epoch": 6.785714285714286,
272
+ "grad_norm": 1.6485190391540527,
273
+ "learning_rate": 8.720000000000001e-06,
274
+ "loss": 0.5115,
275
+ "step": 950
276
+ },
277
+ {
278
+ "epoch": 6.964285714285714,
279
+ "grad_norm": 1.7726221084594727,
280
+ "learning_rate": 8.64857142857143e-06,
281
+ "loss": 0.5183,
282
+ "step": 975
283
+ },
284
+ {
285
+ "epoch": 7.142857142857143,
286
+ "grad_norm": 2.7793469429016113,
287
+ "learning_rate": 8.577142857142858e-06,
288
+ "loss": 0.5133,
289
+ "step": 1000
290
+ },
291
+ {
292
+ "epoch": 7.142857142857143,
293
+ "eval_loss": 0.4698790907859802,
294
+ "eval_runtime": 10.706,
295
+ "eval_samples_per_second": 46.422,
296
+ "eval_steps_per_second": 5.885,
297
+ "step": 1000
298
+ },
299
+ {
300
+ "epoch": 7.321428571428571,
301
+ "grad_norm": 2.8019936084747314,
302
+ "learning_rate": 8.505714285714287e-06,
303
+ "loss": 0.518,
304
+ "step": 1025
305
+ },
306
+ {
307
+ "epoch": 7.5,
308
+ "grad_norm": 2.1961379051208496,
309
+ "learning_rate": 8.434285714285716e-06,
310
+ "loss": 0.5112,
311
+ "step": 1050
312
+ },
313
+ {
314
+ "epoch": 7.678571428571429,
315
+ "grad_norm": 2.5158090591430664,
316
+ "learning_rate": 8.362857142857143e-06,
317
+ "loss": 0.5091,
318
+ "step": 1075
319
+ },
320
+ {
321
+ "epoch": 7.857142857142857,
322
+ "grad_norm": 2.2158660888671875,
323
+ "learning_rate": 8.291428571428572e-06,
324
+ "loss": 0.5121,
325
+ "step": 1100
326
+ },
327
+ {
328
+ "epoch": 8.035714285714286,
329
+ "grad_norm": 2.8917131423950195,
330
+ "learning_rate": 8.220000000000001e-06,
331
+ "loss": 0.5134,
332
+ "step": 1125
333
+ },
334
+ {
335
+ "epoch": 8.214285714285714,
336
+ "grad_norm": 1.9233217239379883,
337
+ "learning_rate": 8.148571428571428e-06,
338
+ "loss": 0.5101,
339
+ "step": 1150
340
+ },
341
+ {
342
+ "epoch": 8.392857142857142,
343
+ "grad_norm": 2.437870979309082,
344
+ "learning_rate": 8.077142857142857e-06,
345
+ "loss": 0.5138,
346
+ "step": 1175
347
+ },
348
+ {
349
+ "epoch": 8.571428571428571,
350
+ "grad_norm": 1.7921634912490845,
351
+ "learning_rate": 8.005714285714286e-06,
352
+ "loss": 0.5155,
353
+ "step": 1200
354
+ },
355
+ {
356
+ "epoch": 8.75,
357
+ "grad_norm": 3.1069226264953613,
358
+ "learning_rate": 7.934285714285715e-06,
359
+ "loss": 0.5144,
360
+ "step": 1225
361
+ },
362
+ {
363
+ "epoch": 8.928571428571429,
364
+ "grad_norm": 1.7902675867080688,
365
+ "learning_rate": 7.862857142857143e-06,
366
+ "loss": 0.4973,
367
+ "step": 1250
368
+ },
369
+ {
370
+ "epoch": 9.107142857142858,
371
+ "grad_norm": 2.353273391723633,
372
+ "learning_rate": 7.791428571428572e-06,
373
+ "loss": 0.5035,
374
+ "step": 1275
375
+ },
376
+ {
377
+ "epoch": 9.285714285714286,
378
+ "grad_norm": 2.3287203311920166,
379
+ "learning_rate": 7.72e-06,
380
+ "loss": 0.5085,
381
+ "step": 1300
382
+ },
383
+ {
384
+ "epoch": 9.464285714285714,
385
+ "grad_norm": 1.6508110761642456,
386
+ "learning_rate": 7.64857142857143e-06,
387
+ "loss": 0.5116,
388
+ "step": 1325
389
+ },
390
+ {
391
+ "epoch": 9.642857142857142,
392
+ "grad_norm": 1.8232289552688599,
393
+ "learning_rate": 7.577142857142857e-06,
394
+ "loss": 0.5033,
395
+ "step": 1350
396
+ },
397
+ {
398
+ "epoch": 9.821428571428571,
399
+ "grad_norm": 2.0248517990112305,
400
+ "learning_rate": 7.505714285714286e-06,
401
+ "loss": 0.4974,
402
+ "step": 1375
403
+ },
404
+ {
405
+ "epoch": 10.0,
406
+ "grad_norm": 3.7484123706817627,
407
+ "learning_rate": 7.434285714285715e-06,
408
+ "loss": 0.4947,
409
+ "step": 1400
410
+ },
411
+ {
412
+ "epoch": 10.178571428571429,
413
+ "grad_norm": 1.793626308441162,
414
+ "learning_rate": 7.362857142857144e-06,
415
+ "loss": 0.5008,
416
+ "step": 1425
417
+ },
418
+ {
419
+ "epoch": 10.357142857142858,
420
+ "grad_norm": 1.8825795650482178,
421
+ "learning_rate": 7.291428571428571e-06,
422
+ "loss": 0.4965,
423
+ "step": 1450
424
+ },
425
+ {
426
+ "epoch": 10.535714285714286,
427
+ "grad_norm": 2.0584166049957275,
428
+ "learning_rate": 7.22e-06,
429
+ "loss": 0.4946,
430
+ "step": 1475
431
+ },
432
+ {
433
+ "epoch": 10.714285714285714,
434
+ "grad_norm": 3.3335680961608887,
435
+ "learning_rate": 7.148571428571429e-06,
436
+ "loss": 0.5048,
437
+ "step": 1500
438
+ },
439
+ {
440
+ "epoch": 10.892857142857142,
441
+ "grad_norm": 1.9972820281982422,
442
+ "learning_rate": 7.077142857142858e-06,
443
+ "loss": 0.5071,
444
+ "step": 1525
445
+ },
446
+ {
447
+ "epoch": 11.071428571428571,
448
+ "grad_norm": 2.2279226779937744,
449
+ "learning_rate": 7.0057142857142865e-06,
450
+ "loss": 0.4874,
451
+ "step": 1550
452
+ },
453
+ {
454
+ "epoch": 11.25,
455
+ "grad_norm": 2.168933868408203,
456
+ "learning_rate": 6.934285714285715e-06,
457
+ "loss": 0.5011,
458
+ "step": 1575
459
+ },
460
+ {
461
+ "epoch": 11.428571428571429,
462
+ "grad_norm": 3.186655044555664,
463
+ "learning_rate": 6.862857142857144e-06,
464
+ "loss": 0.4963,
465
+ "step": 1600
466
+ },
467
+ {
468
+ "epoch": 11.607142857142858,
469
+ "grad_norm": 2.6348812580108643,
470
+ "learning_rate": 6.791428571428572e-06,
471
+ "loss": 0.5007,
472
+ "step": 1625
473
+ },
474
+ {
475
+ "epoch": 11.785714285714286,
476
+ "grad_norm": 3.1159684658050537,
477
+ "learning_rate": 6.720000000000001e-06,
478
+ "loss": 0.4997,
479
+ "step": 1650
480
+ },
481
+ {
482
+ "epoch": 11.964285714285714,
483
+ "grad_norm": 4.068305492401123,
484
+ "learning_rate": 6.648571428571429e-06,
485
+ "loss": 0.4919,
486
+ "step": 1675
487
+ },
488
+ {
489
+ "epoch": 12.142857142857142,
490
+ "grad_norm": 2.315338134765625,
491
+ "learning_rate": 6.577142857142857e-06,
492
+ "loss": 0.4937,
493
+ "step": 1700
494
+ },
495
+ {
496
+ "epoch": 12.321428571428571,
497
+ "grad_norm": 1.5482977628707886,
498
+ "learning_rate": 6.505714285714286e-06,
499
+ "loss": 0.4964,
500
+ "step": 1725
501
+ },
502
+ {
503
+ "epoch": 12.5,
504
+ "grad_norm": 2.1527745723724365,
505
+ "learning_rate": 6.434285714285715e-06,
506
+ "loss": 0.494,
507
+ "step": 1750
508
+ },
509
+ {
510
+ "epoch": 12.678571428571429,
511
+ "grad_norm": 2.216703176498413,
512
+ "learning_rate": 6.3628571428571426e-06,
513
+ "loss": 0.4853,
514
+ "step": 1775
515
+ },
516
+ {
517
+ "epoch": 12.857142857142858,
518
+ "grad_norm": 2.939267873764038,
519
+ "learning_rate": 6.2914285714285716e-06,
520
+ "loss": 0.4978,
521
+ "step": 1800
522
+ },
523
+ {
524
+ "epoch": 13.035714285714286,
525
+ "grad_norm": 2.3067381381988525,
526
+ "learning_rate": 6.220000000000001e-06,
527
+ "loss": 0.5023,
528
+ "step": 1825
529
+ },
530
+ {
531
+ "epoch": 13.214285714285714,
532
+ "grad_norm": 2.2849552631378174,
533
+ "learning_rate": 6.14857142857143e-06,
534
+ "loss": 0.5013,
535
+ "step": 1850
536
+ },
537
+ {
538
+ "epoch": 13.392857142857142,
539
+ "grad_norm": 1.8568611145019531,
540
+ "learning_rate": 6.077142857142858e-06,
541
+ "loss": 0.4855,
542
+ "step": 1875
543
+ },
544
+ {
545
+ "epoch": 13.571428571428571,
546
+ "grad_norm": 2.784388542175293,
547
+ "learning_rate": 6.005714285714286e-06,
548
+ "loss": 0.4832,
549
+ "step": 1900
550
+ },
551
+ {
552
+ "epoch": 13.75,
553
+ "grad_norm": 1.718732237815857,
554
+ "learning_rate": 5.934285714285715e-06,
555
+ "loss": 0.4926,
556
+ "step": 1925
557
+ },
558
+ {
559
+ "epoch": 13.928571428571429,
560
+ "grad_norm": 2.3010714054107666,
561
+ "learning_rate": 5.862857142857143e-06,
562
+ "loss": 0.494,
563
+ "step": 1950
564
+ },
565
+ {
566
+ "epoch": 14.107142857142858,
567
+ "grad_norm": 2.0058493614196777,
568
+ "learning_rate": 5.791428571428572e-06,
569
+ "loss": 0.4901,
570
+ "step": 1975
571
+ },
572
+ {
573
+ "epoch": 14.285714285714286,
574
+ "grad_norm": 2.311893939971924,
575
+ "learning_rate": 5.72e-06,
576
+ "loss": 0.4774,
577
+ "step": 2000
578
+ },
579
+ {
580
+ "epoch": 14.285714285714286,
581
+ "eval_loss": 0.45390585064888,
582
+ "eval_runtime": 18.3883,
583
+ "eval_samples_per_second": 27.028,
584
+ "eval_steps_per_second": 3.426,
585
+ "step": 2000
586
+ }
587
+ ],
588
+ "logging_steps": 25,
589
+ "max_steps": 4000,
590
+ "num_input_tokens_seen": 0,
591
+ "num_train_epochs": 29,
592
+ "save_steps": 1000,
593
+ "stateful_callbacks": {
594
+ "TrainerControl": {
595
+ "args": {
596
+ "should_epoch_stop": false,
597
+ "should_evaluate": false,
598
+ "should_log": false,
599
+ "should_save": true,
600
+ "should_training_stop": false
601
+ },
602
+ "attributes": {}
603
+ }
604
+ },
605
+ "total_flos": 5927807866030776.0,
606
+ "train_batch_size": 16,
607
+ "trial_name": null,
608
+ "trial_params": null
609
+ }
checkpoint-2000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a01e8c0a090b02699d5b506a30ffaf176234c63870a5a741f73db593b385337
3
+ size 5560
checkpoint-3000/added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "<ctc_blank>": 80,
3
+ "<mask>": 79
4
+ }
checkpoint-3000/config.json ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/speecht5_tts",
3
+ "activation_dropout": 0.1,
4
+ "apply_spec_augment": true,
5
+ "architectures": [
6
+ "SpeechT5ForTextToSpeech"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 0,
10
+ "conv_bias": false,
11
+ "conv_dim": [
12
+ 512,
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512
19
+ ],
20
+ "conv_kernel": [
21
+ 10,
22
+ 3,
23
+ 3,
24
+ 3,
25
+ 3,
26
+ 2,
27
+ 2
28
+ ],
29
+ "conv_stride": [
30
+ 5,
31
+ 2,
32
+ 2,
33
+ 2,
34
+ 2,
35
+ 2,
36
+ 2
37
+ ],
38
+ "decoder_attention_heads": 12,
39
+ "decoder_ffn_dim": 3072,
40
+ "decoder_layerdrop": 0.1,
41
+ "decoder_layers": 6,
42
+ "decoder_start_token_id": 2,
43
+ "encoder_attention_heads": 12,
44
+ "encoder_ffn_dim": 3072,
45
+ "encoder_layerdrop": 0.1,
46
+ "encoder_layers": 12,
47
+ "encoder_max_relative_position": 160,
48
+ "eos_token_id": 2,
49
+ "feat_extract_activation": "gelu",
50
+ "feat_extract_norm": "group",
51
+ "feat_proj_dropout": 0.0,
52
+ "guided_attention_loss_num_heads": 2,
53
+ "guided_attention_loss_scale": 10.0,
54
+ "guided_attention_loss_sigma": 0.4,
55
+ "hidden_act": "gelu",
56
+ "hidden_dropout": 0.1,
57
+ "hidden_size": 768,
58
+ "initializer_range": 0.02,
59
+ "is_encoder_decoder": true,
60
+ "layer_norm_eps": 1e-05,
61
+ "mask_feature_length": 10,
62
+ "mask_feature_min_masks": 0,
63
+ "mask_feature_prob": 0.0,
64
+ "mask_time_length": 10,
65
+ "mask_time_min_masks": 2,
66
+ "mask_time_prob": 0.05,
67
+ "max_length": 1876,
68
+ "max_speech_positions": 1876,
69
+ "max_text_positions": 600,
70
+ "model_type": "speecht5",
71
+ "num_conv_pos_embedding_groups": 16,
72
+ "num_conv_pos_embeddings": 128,
73
+ "num_feat_extract_layers": 7,
74
+ "num_mel_bins": 80,
75
+ "pad_token_id": 1,
76
+ "positional_dropout": 0.1,
77
+ "reduction_factor": 2,
78
+ "scale_embedding": false,
79
+ "speaker_embedding_dim": 512,
80
+ "speech_decoder_postnet_dropout": 0.5,
81
+ "speech_decoder_postnet_kernel": 5,
82
+ "speech_decoder_postnet_layers": 5,
83
+ "speech_decoder_postnet_units": 256,
84
+ "speech_decoder_prenet_dropout": 0.5,
85
+ "speech_decoder_prenet_layers": 2,
86
+ "speech_decoder_prenet_units": 256,
87
+ "torch_dtype": "float32",
88
+ "transformers_version": "4.50.0.dev0",
89
+ "use_cache": false,
90
+ "use_guided_attention_loss": true,
91
+ "vocab_size": 81
92
+ }
checkpoint-3000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1cd68e450899c334ac78ad0994af7cabde4187c89fe02b7215159c76c2bce334
3
+ size 577789320
checkpoint-3000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c87eef7f0e2e00cbab7038d913581ba1792c5c003d283732899044684a57c635
3
+ size 1155772233
checkpoint-3000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ebd1aefed45469c63901268b01ba1869f235267cd8c436935a74b257cb260841
3
+ size 14244
checkpoint-3000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a73b49b55d3aadf9516bb5b6bdc1d8930fb0d3a40da81083b6f1d89b5130278c
3
+ size 988
checkpoint-3000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:30b4b82462a87716b3f11ac99d935386eaf7ea7c7ab13974e514e9689621e7eb
3
+ size 1064
checkpoint-3000/special_tokens_map.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "eos_token": "</s>",
4
+ "mask_token": {
5
+ "content": "<mask>",
6
+ "lstrip": true,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false
10
+ },
11
+ "pad_token": "<pad>",
12
+ "unk_token": "<unk>"
13
+ }
checkpoint-3000/spm_char.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7fcc48f3e225f627b1641db410ceb0c8649bd2b0c982e150b03f8be3728ab560
3
+ size 238473
checkpoint-3000/tokenizer_config.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "79": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": true,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "80": {
44
+ "content": "<ctc_blank>",
45
+ "lstrip": false,
46
+ "normalized": true,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": false
50
+ }
51
+ },
52
+ "bos_token": "<s>",
53
+ "clean_up_tokenization_spaces": false,
54
+ "eos_token": "</s>",
55
+ "extra_special_tokens": {},
56
+ "mask_token": "<mask>",
57
+ "model_max_length": 600,
58
+ "normalize": false,
59
+ "pad_token": "<pad>",
60
+ "processor_class": "SpeechT5Processor",
61
+ "sp_model_kwargs": {},
62
+ "tokenizer_class": "SpeechT5Tokenizer",
63
+ "unk_token": "<unk>"
64
+ }
checkpoint-3000/trainer_state.json ADDED
@@ -0,0 +1,897 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.44868698716163635,
3
+ "best_model_checkpoint": "./speecht5_tts_common_voice_17_0_id/checkpoint-3000",
4
+ "epoch": 21.428571428571427,
5
+ "eval_steps": 1000,
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": 0.17857142857142858,
13
+ "grad_norm": 4.796061038970947,
14
+ "learning_rate": 4.6000000000000004e-07,
15
+ "loss": 1.0232,
16
+ "step": 25
17
+ },
18
+ {
19
+ "epoch": 0.35714285714285715,
20
+ "grad_norm": 6.88640022277832,
21
+ "learning_rate": 9.600000000000001e-07,
22
+ "loss": 0.9412,
23
+ "step": 50
24
+ },
25
+ {
26
+ "epoch": 0.5357142857142857,
27
+ "grad_norm": 4.282207489013672,
28
+ "learning_rate": 1.46e-06,
29
+ "loss": 0.8431,
30
+ "step": 75
31
+ },
32
+ {
33
+ "epoch": 0.7142857142857143,
34
+ "grad_norm": 4.095445156097412,
35
+ "learning_rate": 1.9600000000000003e-06,
36
+ "loss": 0.7959,
37
+ "step": 100
38
+ },
39
+ {
40
+ "epoch": 0.8928571428571429,
41
+ "grad_norm": 6.723930835723877,
42
+ "learning_rate": 2.46e-06,
43
+ "loss": 0.7825,
44
+ "step": 125
45
+ },
46
+ {
47
+ "epoch": 1.0714285714285714,
48
+ "grad_norm": 2.318833589553833,
49
+ "learning_rate": 2.96e-06,
50
+ "loss": 0.7967,
51
+ "step": 150
52
+ },
53
+ {
54
+ "epoch": 1.25,
55
+ "grad_norm": 2.1883444786071777,
56
+ "learning_rate": 3.46e-06,
57
+ "loss": 0.7826,
58
+ "step": 175
59
+ },
60
+ {
61
+ "epoch": 1.4285714285714286,
62
+ "grad_norm": 2.7793357372283936,
63
+ "learning_rate": 3.96e-06,
64
+ "loss": 0.7238,
65
+ "step": 200
66
+ },
67
+ {
68
+ "epoch": 1.6071428571428572,
69
+ "grad_norm": 3.0292282104492188,
70
+ "learning_rate": 4.4600000000000005e-06,
71
+ "loss": 0.6981,
72
+ "step": 225
73
+ },
74
+ {
75
+ "epoch": 1.7857142857142856,
76
+ "grad_norm": 4.8227314949035645,
77
+ "learning_rate": 4.960000000000001e-06,
78
+ "loss": 0.6753,
79
+ "step": 250
80
+ },
81
+ {
82
+ "epoch": 1.9642857142857144,
83
+ "grad_norm": 2.781235694885254,
84
+ "learning_rate": 5.460000000000001e-06,
85
+ "loss": 0.6337,
86
+ "step": 275
87
+ },
88
+ {
89
+ "epoch": 2.142857142857143,
90
+ "grad_norm": 2.9240171909332275,
91
+ "learning_rate": 5.9600000000000005e-06,
92
+ "loss": 0.6499,
93
+ "step": 300
94
+ },
95
+ {
96
+ "epoch": 2.3214285714285716,
97
+ "grad_norm": 2.0720555782318115,
98
+ "learning_rate": 6.460000000000001e-06,
99
+ "loss": 0.5995,
100
+ "step": 325
101
+ },
102
+ {
103
+ "epoch": 2.5,
104
+ "grad_norm": 3.025918483734131,
105
+ "learning_rate": 6.96e-06,
106
+ "loss": 0.6054,
107
+ "step": 350
108
+ },
109
+ {
110
+ "epoch": 2.678571428571429,
111
+ "grad_norm": 3.3047034740448,
112
+ "learning_rate": 7.4600000000000006e-06,
113
+ "loss": 0.5828,
114
+ "step": 375
115
+ },
116
+ {
117
+ "epoch": 2.857142857142857,
118
+ "grad_norm": 2.4087064266204834,
119
+ "learning_rate": 7.960000000000002e-06,
120
+ "loss": 0.5706,
121
+ "step": 400
122
+ },
123
+ {
124
+ "epoch": 3.0357142857142856,
125
+ "grad_norm": 2.7837984561920166,
126
+ "learning_rate": 8.46e-06,
127
+ "loss": 0.579,
128
+ "step": 425
129
+ },
130
+ {
131
+ "epoch": 3.2142857142857144,
132
+ "grad_norm": 1.7530593872070312,
133
+ "learning_rate": 8.96e-06,
134
+ "loss": 0.5455,
135
+ "step": 450
136
+ },
137
+ {
138
+ "epoch": 3.392857142857143,
139
+ "grad_norm": 2.3677515983581543,
140
+ "learning_rate": 9.460000000000001e-06,
141
+ "loss": 0.5557,
142
+ "step": 475
143
+ },
144
+ {
145
+ "epoch": 3.571428571428571,
146
+ "grad_norm": 2.445343494415283,
147
+ "learning_rate": 9.960000000000001e-06,
148
+ "loss": 0.5686,
149
+ "step": 500
150
+ },
151
+ {
152
+ "epoch": 3.75,
153
+ "grad_norm": 4.732877731323242,
154
+ "learning_rate": 9.934285714285715e-06,
155
+ "loss": 0.5635,
156
+ "step": 525
157
+ },
158
+ {
159
+ "epoch": 3.928571428571429,
160
+ "grad_norm": 4.45880126953125,
161
+ "learning_rate": 9.862857142857144e-06,
162
+ "loss": 0.5477,
163
+ "step": 550
164
+ },
165
+ {
166
+ "epoch": 4.107142857142857,
167
+ "grad_norm": 3.9657559394836426,
168
+ "learning_rate": 9.791428571428571e-06,
169
+ "loss": 0.5563,
170
+ "step": 575
171
+ },
172
+ {
173
+ "epoch": 4.285714285714286,
174
+ "grad_norm": 2.676309823989868,
175
+ "learning_rate": 9.72e-06,
176
+ "loss": 0.5221,
177
+ "step": 600
178
+ },
179
+ {
180
+ "epoch": 4.464285714285714,
181
+ "grad_norm": 2.89937686920166,
182
+ "learning_rate": 9.648571428571429e-06,
183
+ "loss": 0.5211,
184
+ "step": 625
185
+ },
186
+ {
187
+ "epoch": 4.642857142857143,
188
+ "grad_norm": 2.9824204444885254,
189
+ "learning_rate": 9.577142857142858e-06,
190
+ "loss": 0.5583,
191
+ "step": 650
192
+ },
193
+ {
194
+ "epoch": 4.821428571428571,
195
+ "grad_norm": 3.632369041442871,
196
+ "learning_rate": 9.505714285714287e-06,
197
+ "loss": 0.526,
198
+ "step": 675
199
+ },
200
+ {
201
+ "epoch": 5.0,
202
+ "grad_norm": 2.352837324142456,
203
+ "learning_rate": 9.434285714285714e-06,
204
+ "loss": 0.5211,
205
+ "step": 700
206
+ },
207
+ {
208
+ "epoch": 5.178571428571429,
209
+ "grad_norm": 1.3236348628997803,
210
+ "learning_rate": 9.362857142857143e-06,
211
+ "loss": 0.5173,
212
+ "step": 725
213
+ },
214
+ {
215
+ "epoch": 5.357142857142857,
216
+ "grad_norm": 1.8493444919586182,
217
+ "learning_rate": 9.291428571428572e-06,
218
+ "loss": 0.5094,
219
+ "step": 750
220
+ },
221
+ {
222
+ "epoch": 5.535714285714286,
223
+ "grad_norm": 1.7403544187545776,
224
+ "learning_rate": 9.220000000000002e-06,
225
+ "loss": 0.5239,
226
+ "step": 775
227
+ },
228
+ {
229
+ "epoch": 5.714285714285714,
230
+ "grad_norm": 4.10603666305542,
231
+ "learning_rate": 9.148571428571429e-06,
232
+ "loss": 0.5304,
233
+ "step": 800
234
+ },
235
+ {
236
+ "epoch": 5.892857142857143,
237
+ "grad_norm": 2.69100284576416,
238
+ "learning_rate": 9.077142857142858e-06,
239
+ "loss": 0.5224,
240
+ "step": 825
241
+ },
242
+ {
243
+ "epoch": 6.071428571428571,
244
+ "grad_norm": 1.956499695777893,
245
+ "learning_rate": 9.005714285714287e-06,
246
+ "loss": 0.5065,
247
+ "step": 850
248
+ },
249
+ {
250
+ "epoch": 6.25,
251
+ "grad_norm": 1.915540337562561,
252
+ "learning_rate": 8.934285714285716e-06,
253
+ "loss": 0.511,
254
+ "step": 875
255
+ },
256
+ {
257
+ "epoch": 6.428571428571429,
258
+ "grad_norm": 2.1903812885284424,
259
+ "learning_rate": 8.862857142857143e-06,
260
+ "loss": 0.5132,
261
+ "step": 900
262
+ },
263
+ {
264
+ "epoch": 6.607142857142857,
265
+ "grad_norm": 1.666169285774231,
266
+ "learning_rate": 8.791428571428572e-06,
267
+ "loss": 0.5362,
268
+ "step": 925
269
+ },
270
+ {
271
+ "epoch": 6.785714285714286,
272
+ "grad_norm": 1.6485190391540527,
273
+ "learning_rate": 8.720000000000001e-06,
274
+ "loss": 0.5115,
275
+ "step": 950
276
+ },
277
+ {
278
+ "epoch": 6.964285714285714,
279
+ "grad_norm": 1.7726221084594727,
280
+ "learning_rate": 8.64857142857143e-06,
281
+ "loss": 0.5183,
282
+ "step": 975
283
+ },
284
+ {
285
+ "epoch": 7.142857142857143,
286
+ "grad_norm": 2.7793469429016113,
287
+ "learning_rate": 8.577142857142858e-06,
288
+ "loss": 0.5133,
289
+ "step": 1000
290
+ },
291
+ {
292
+ "epoch": 7.142857142857143,
293
+ "eval_loss": 0.4698790907859802,
294
+ "eval_runtime": 10.706,
295
+ "eval_samples_per_second": 46.422,
296
+ "eval_steps_per_second": 5.885,
297
+ "step": 1000
298
+ },
299
+ {
300
+ "epoch": 7.321428571428571,
301
+ "grad_norm": 2.8019936084747314,
302
+ "learning_rate": 8.505714285714287e-06,
303
+ "loss": 0.518,
304
+ "step": 1025
305
+ },
306
+ {
307
+ "epoch": 7.5,
308
+ "grad_norm": 2.1961379051208496,
309
+ "learning_rate": 8.434285714285716e-06,
310
+ "loss": 0.5112,
311
+ "step": 1050
312
+ },
313
+ {
314
+ "epoch": 7.678571428571429,
315
+ "grad_norm": 2.5158090591430664,
316
+ "learning_rate": 8.362857142857143e-06,
317
+ "loss": 0.5091,
318
+ "step": 1075
319
+ },
320
+ {
321
+ "epoch": 7.857142857142857,
322
+ "grad_norm": 2.2158660888671875,
323
+ "learning_rate": 8.291428571428572e-06,
324
+ "loss": 0.5121,
325
+ "step": 1100
326
+ },
327
+ {
328
+ "epoch": 8.035714285714286,
329
+ "grad_norm": 2.8917131423950195,
330
+ "learning_rate": 8.220000000000001e-06,
331
+ "loss": 0.5134,
332
+ "step": 1125
333
+ },
334
+ {
335
+ "epoch": 8.214285714285714,
336
+ "grad_norm": 1.9233217239379883,
337
+ "learning_rate": 8.148571428571428e-06,
338
+ "loss": 0.5101,
339
+ "step": 1150
340
+ },
341
+ {
342
+ "epoch": 8.392857142857142,
343
+ "grad_norm": 2.437870979309082,
344
+ "learning_rate": 8.077142857142857e-06,
345
+ "loss": 0.5138,
346
+ "step": 1175
347
+ },
348
+ {
349
+ "epoch": 8.571428571428571,
350
+ "grad_norm": 1.7921634912490845,
351
+ "learning_rate": 8.005714285714286e-06,
352
+ "loss": 0.5155,
353
+ "step": 1200
354
+ },
355
+ {
356
+ "epoch": 8.75,
357
+ "grad_norm": 3.1069226264953613,
358
+ "learning_rate": 7.934285714285715e-06,
359
+ "loss": 0.5144,
360
+ "step": 1225
361
+ },
362
+ {
363
+ "epoch": 8.928571428571429,
364
+ "grad_norm": 1.7902675867080688,
365
+ "learning_rate": 7.862857142857143e-06,
366
+ "loss": 0.4973,
367
+ "step": 1250
368
+ },
369
+ {
370
+ "epoch": 9.107142857142858,
371
+ "grad_norm": 2.353273391723633,
372
+ "learning_rate": 7.791428571428572e-06,
373
+ "loss": 0.5035,
374
+ "step": 1275
375
+ },
376
+ {
377
+ "epoch": 9.285714285714286,
378
+ "grad_norm": 2.3287203311920166,
379
+ "learning_rate": 7.72e-06,
380
+ "loss": 0.5085,
381
+ "step": 1300
382
+ },
383
+ {
384
+ "epoch": 9.464285714285714,
385
+ "grad_norm": 1.6508110761642456,
386
+ "learning_rate": 7.64857142857143e-06,
387
+ "loss": 0.5116,
388
+ "step": 1325
389
+ },
390
+ {
391
+ "epoch": 9.642857142857142,
392
+ "grad_norm": 1.8232289552688599,
393
+ "learning_rate": 7.577142857142857e-06,
394
+ "loss": 0.5033,
395
+ "step": 1350
396
+ },
397
+ {
398
+ "epoch": 9.821428571428571,
399
+ "grad_norm": 2.0248517990112305,
400
+ "learning_rate": 7.505714285714286e-06,
401
+ "loss": 0.4974,
402
+ "step": 1375
403
+ },
404
+ {
405
+ "epoch": 10.0,
406
+ "grad_norm": 3.7484123706817627,
407
+ "learning_rate": 7.434285714285715e-06,
408
+ "loss": 0.4947,
409
+ "step": 1400
410
+ },
411
+ {
412
+ "epoch": 10.178571428571429,
413
+ "grad_norm": 1.793626308441162,
414
+ "learning_rate": 7.362857142857144e-06,
415
+ "loss": 0.5008,
416
+ "step": 1425
417
+ },
418
+ {
419
+ "epoch": 10.357142857142858,
420
+ "grad_norm": 1.8825795650482178,
421
+ "learning_rate": 7.291428571428571e-06,
422
+ "loss": 0.4965,
423
+ "step": 1450
424
+ },
425
+ {
426
+ "epoch": 10.535714285714286,
427
+ "grad_norm": 2.0584166049957275,
428
+ "learning_rate": 7.22e-06,
429
+ "loss": 0.4946,
430
+ "step": 1475
431
+ },
432
+ {
433
+ "epoch": 10.714285714285714,
434
+ "grad_norm": 3.3335680961608887,
435
+ "learning_rate": 7.148571428571429e-06,
436
+ "loss": 0.5048,
437
+ "step": 1500
438
+ },
439
+ {
440
+ "epoch": 10.892857142857142,
441
+ "grad_norm": 1.9972820281982422,
442
+ "learning_rate": 7.077142857142858e-06,
443
+ "loss": 0.5071,
444
+ "step": 1525
445
+ },
446
+ {
447
+ "epoch": 11.071428571428571,
448
+ "grad_norm": 2.2279226779937744,
449
+ "learning_rate": 7.0057142857142865e-06,
450
+ "loss": 0.4874,
451
+ "step": 1550
452
+ },
453
+ {
454
+ "epoch": 11.25,
455
+ "grad_norm": 2.168933868408203,
456
+ "learning_rate": 6.934285714285715e-06,
457
+ "loss": 0.5011,
458
+ "step": 1575
459
+ },
460
+ {
461
+ "epoch": 11.428571428571429,
462
+ "grad_norm": 3.186655044555664,
463
+ "learning_rate": 6.862857142857144e-06,
464
+ "loss": 0.4963,
465
+ "step": 1600
466
+ },
467
+ {
468
+ "epoch": 11.607142857142858,
469
+ "grad_norm": 2.6348812580108643,
470
+ "learning_rate": 6.791428571428572e-06,
471
+ "loss": 0.5007,
472
+ "step": 1625
473
+ },
474
+ {
475
+ "epoch": 11.785714285714286,
476
+ "grad_norm": 3.1159684658050537,
477
+ "learning_rate": 6.720000000000001e-06,
478
+ "loss": 0.4997,
479
+ "step": 1650
480
+ },
481
+ {
482
+ "epoch": 11.964285714285714,
483
+ "grad_norm": 4.068305492401123,
484
+ "learning_rate": 6.648571428571429e-06,
485
+ "loss": 0.4919,
486
+ "step": 1675
487
+ },
488
+ {
489
+ "epoch": 12.142857142857142,
490
+ "grad_norm": 2.315338134765625,
491
+ "learning_rate": 6.577142857142857e-06,
492
+ "loss": 0.4937,
493
+ "step": 1700
494
+ },
495
+ {
496
+ "epoch": 12.321428571428571,
497
+ "grad_norm": 1.5482977628707886,
498
+ "learning_rate": 6.505714285714286e-06,
499
+ "loss": 0.4964,
500
+ "step": 1725
501
+ },
502
+ {
503
+ "epoch": 12.5,
504
+ "grad_norm": 2.1527745723724365,
505
+ "learning_rate": 6.434285714285715e-06,
506
+ "loss": 0.494,
507
+ "step": 1750
508
+ },
509
+ {
510
+ "epoch": 12.678571428571429,
511
+ "grad_norm": 2.216703176498413,
512
+ "learning_rate": 6.3628571428571426e-06,
513
+ "loss": 0.4853,
514
+ "step": 1775
515
+ },
516
+ {
517
+ "epoch": 12.857142857142858,
518
+ "grad_norm": 2.939267873764038,
519
+ "learning_rate": 6.2914285714285716e-06,
520
+ "loss": 0.4978,
521
+ "step": 1800
522
+ },
523
+ {
524
+ "epoch": 13.035714285714286,
525
+ "grad_norm": 2.3067381381988525,
526
+ "learning_rate": 6.220000000000001e-06,
527
+ "loss": 0.5023,
528
+ "step": 1825
529
+ },
530
+ {
531
+ "epoch": 13.214285714285714,
532
+ "grad_norm": 2.2849552631378174,
533
+ "learning_rate": 6.14857142857143e-06,
534
+ "loss": 0.5013,
535
+ "step": 1850
536
+ },
537
+ {
538
+ "epoch": 13.392857142857142,
539
+ "grad_norm": 1.8568611145019531,
540
+ "learning_rate": 6.077142857142858e-06,
541
+ "loss": 0.4855,
542
+ "step": 1875
543
+ },
544
+ {
545
+ "epoch": 13.571428571428571,
546
+ "grad_norm": 2.784388542175293,
547
+ "learning_rate": 6.005714285714286e-06,
548
+ "loss": 0.4832,
549
+ "step": 1900
550
+ },
551
+ {
552
+ "epoch": 13.75,
553
+ "grad_norm": 1.718732237815857,
554
+ "learning_rate": 5.934285714285715e-06,
555
+ "loss": 0.4926,
556
+ "step": 1925
557
+ },
558
+ {
559
+ "epoch": 13.928571428571429,
560
+ "grad_norm": 2.3010714054107666,
561
+ "learning_rate": 5.862857142857143e-06,
562
+ "loss": 0.494,
563
+ "step": 1950
564
+ },
565
+ {
566
+ "epoch": 14.107142857142858,
567
+ "grad_norm": 2.0058493614196777,
568
+ "learning_rate": 5.791428571428572e-06,
569
+ "loss": 0.4901,
570
+ "step": 1975
571
+ },
572
+ {
573
+ "epoch": 14.285714285714286,
574
+ "grad_norm": 2.311893939971924,
575
+ "learning_rate": 5.72e-06,
576
+ "loss": 0.4774,
577
+ "step": 2000
578
+ },
579
+ {
580
+ "epoch": 14.285714285714286,
581
+ "eval_loss": 0.45390585064888,
582
+ "eval_runtime": 18.3883,
583
+ "eval_samples_per_second": 27.028,
584
+ "eval_steps_per_second": 3.426,
585
+ "step": 2000
586
+ },
587
+ {
588
+ "epoch": 14.464285714285714,
589
+ "grad_norm": 5.475688457489014,
590
+ "learning_rate": 5.6485714285714285e-06,
591
+ "loss": 0.479,
592
+ "step": 2025
593
+ },
594
+ {
595
+ "epoch": 14.642857142857142,
596
+ "grad_norm": 2.333171844482422,
597
+ "learning_rate": 5.5771428571428575e-06,
598
+ "loss": 0.487,
599
+ "step": 2050
600
+ },
601
+ {
602
+ "epoch": 14.821428571428571,
603
+ "grad_norm": 1.7914478778839111,
604
+ "learning_rate": 5.5057142857142865e-06,
605
+ "loss": 0.5042,
606
+ "step": 2075
607
+ },
608
+ {
609
+ "epoch": 15.0,
610
+ "grad_norm": 2.343949317932129,
611
+ "learning_rate": 5.4342857142857155e-06,
612
+ "loss": 0.4994,
613
+ "step": 2100
614
+ },
615
+ {
616
+ "epoch": 15.178571428571429,
617
+ "grad_norm": 2.7536513805389404,
618
+ "learning_rate": 5.362857142857143e-06,
619
+ "loss": 0.492,
620
+ "step": 2125
621
+ },
622
+ {
623
+ "epoch": 15.357142857142858,
624
+ "grad_norm": 3.5288944244384766,
625
+ "learning_rate": 5.291428571428572e-06,
626
+ "loss": 0.4878,
627
+ "step": 2150
628
+ },
629
+ {
630
+ "epoch": 15.535714285714286,
631
+ "grad_norm": 4.280928611755371,
632
+ "learning_rate": 5.220000000000001e-06,
633
+ "loss": 0.4908,
634
+ "step": 2175
635
+ },
636
+ {
637
+ "epoch": 15.714285714285714,
638
+ "grad_norm": 3.1674723625183105,
639
+ "learning_rate": 5.14857142857143e-06,
640
+ "loss": 0.4896,
641
+ "step": 2200
642
+ },
643
+ {
644
+ "epoch": 15.892857142857142,
645
+ "grad_norm": 1.4766712188720703,
646
+ "learning_rate": 5.077142857142857e-06,
647
+ "loss": 0.4891,
648
+ "step": 2225
649
+ },
650
+ {
651
+ "epoch": 16.071428571428573,
652
+ "grad_norm": 2.7163197994232178,
653
+ "learning_rate": 5.005714285714286e-06,
654
+ "loss": 0.4901,
655
+ "step": 2250
656
+ },
657
+ {
658
+ "epoch": 16.25,
659
+ "grad_norm": 2.2624833583831787,
660
+ "learning_rate": 4.934285714285715e-06,
661
+ "loss": 0.4841,
662
+ "step": 2275
663
+ },
664
+ {
665
+ "epoch": 16.428571428571427,
666
+ "grad_norm": 3.0063624382019043,
667
+ "learning_rate": 4.862857142857143e-06,
668
+ "loss": 0.4826,
669
+ "step": 2300
670
+ },
671
+ {
672
+ "epoch": 16.607142857142858,
673
+ "grad_norm": 1.5605120658874512,
674
+ "learning_rate": 4.7914285714285715e-06,
675
+ "loss": 0.4808,
676
+ "step": 2325
677
+ },
678
+ {
679
+ "epoch": 16.785714285714285,
680
+ "grad_norm": 1.8184423446655273,
681
+ "learning_rate": 4.7200000000000005e-06,
682
+ "loss": 0.4743,
683
+ "step": 2350
684
+ },
685
+ {
686
+ "epoch": 16.964285714285715,
687
+ "grad_norm": 1.490218162536621,
688
+ "learning_rate": 4.648571428571429e-06,
689
+ "loss": 0.4848,
690
+ "step": 2375
691
+ },
692
+ {
693
+ "epoch": 17.142857142857142,
694
+ "grad_norm": 2.210409641265869,
695
+ "learning_rate": 4.577142857142858e-06,
696
+ "loss": 0.4814,
697
+ "step": 2400
698
+ },
699
+ {
700
+ "epoch": 17.321428571428573,
701
+ "grad_norm": 1.8189564943313599,
702
+ "learning_rate": 4.505714285714286e-06,
703
+ "loss": 0.4749,
704
+ "step": 2425
705
+ },
706
+ {
707
+ "epoch": 17.5,
708
+ "grad_norm": 2.2493181228637695,
709
+ "learning_rate": 4.434285714285715e-06,
710
+ "loss": 0.4835,
711
+ "step": 2450
712
+ },
713
+ {
714
+ "epoch": 17.678571428571427,
715
+ "grad_norm": 2.454582452774048,
716
+ "learning_rate": 4.362857142857143e-06,
717
+ "loss": 0.4891,
718
+ "step": 2475
719
+ },
720
+ {
721
+ "epoch": 17.857142857142858,
722
+ "grad_norm": 2.0914249420166016,
723
+ "learning_rate": 4.291428571428572e-06,
724
+ "loss": 0.4774,
725
+ "step": 2500
726
+ },
727
+ {
728
+ "epoch": 18.035714285714285,
729
+ "grad_norm": 1.6150003671646118,
730
+ "learning_rate": 4.22e-06,
731
+ "loss": 0.4826,
732
+ "step": 2525
733
+ },
734
+ {
735
+ "epoch": 18.214285714285715,
736
+ "grad_norm": 2.94893217086792,
737
+ "learning_rate": 4.148571428571429e-06,
738
+ "loss": 0.4894,
739
+ "step": 2550
740
+ },
741
+ {
742
+ "epoch": 18.392857142857142,
743
+ "grad_norm": 1.913024663925171,
744
+ "learning_rate": 4.0771428571428574e-06,
745
+ "loss": 0.4784,
746
+ "step": 2575
747
+ },
748
+ {
749
+ "epoch": 18.571428571428573,
750
+ "grad_norm": 2.2716126441955566,
751
+ "learning_rate": 4.0057142857142864e-06,
752
+ "loss": 0.4709,
753
+ "step": 2600
754
+ },
755
+ {
756
+ "epoch": 18.75,
757
+ "grad_norm": 3.5856099128723145,
758
+ "learning_rate": 3.934285714285715e-06,
759
+ "loss": 0.4885,
760
+ "step": 2625
761
+ },
762
+ {
763
+ "epoch": 18.928571428571427,
764
+ "grad_norm": 4.390285015106201,
765
+ "learning_rate": 3.862857142857143e-06,
766
+ "loss": 0.4828,
767
+ "step": 2650
768
+ },
769
+ {
770
+ "epoch": 19.107142857142858,
771
+ "grad_norm": 3.690558910369873,
772
+ "learning_rate": 3.7914285714285722e-06,
773
+ "loss": 0.4862,
774
+ "step": 2675
775
+ },
776
+ {
777
+ "epoch": 19.285714285714285,
778
+ "grad_norm": 3.1892879009246826,
779
+ "learning_rate": 3.7200000000000004e-06,
780
+ "loss": 0.4798,
781
+ "step": 2700
782
+ },
783
+ {
784
+ "epoch": 19.464285714285715,
785
+ "grad_norm": 2.4511208534240723,
786
+ "learning_rate": 3.648571428571429e-06,
787
+ "loss": 0.4844,
788
+ "step": 2725
789
+ },
790
+ {
791
+ "epoch": 19.642857142857142,
792
+ "grad_norm": 1.642425537109375,
793
+ "learning_rate": 3.5771428571428576e-06,
794
+ "loss": 0.4845,
795
+ "step": 2750
796
+ },
797
+ {
798
+ "epoch": 19.821428571428573,
799
+ "grad_norm": 1.53347909450531,
800
+ "learning_rate": 3.505714285714286e-06,
801
+ "loss": 0.4821,
802
+ "step": 2775
803
+ },
804
+ {
805
+ "epoch": 20.0,
806
+ "grad_norm": 2.1212124824523926,
807
+ "learning_rate": 3.4342857142857143e-06,
808
+ "loss": 0.4756,
809
+ "step": 2800
810
+ },
811
+ {
812
+ "epoch": 20.178571428571427,
813
+ "grad_norm": 2.1033811569213867,
814
+ "learning_rate": 3.3628571428571433e-06,
815
+ "loss": 0.4729,
816
+ "step": 2825
817
+ },
818
+ {
819
+ "epoch": 20.357142857142858,
820
+ "grad_norm": 1.9785221815109253,
821
+ "learning_rate": 3.2914285714285715e-06,
822
+ "loss": 0.4783,
823
+ "step": 2850
824
+ },
825
+ {
826
+ "epoch": 20.535714285714285,
827
+ "grad_norm": 1.7057210206985474,
828
+ "learning_rate": 3.2200000000000005e-06,
829
+ "loss": 0.4728,
830
+ "step": 2875
831
+ },
832
+ {
833
+ "epoch": 20.714285714285715,
834
+ "grad_norm": 2.1187682151794434,
835
+ "learning_rate": 3.1485714285714287e-06,
836
+ "loss": 0.4765,
837
+ "step": 2900
838
+ },
839
+ {
840
+ "epoch": 20.892857142857142,
841
+ "grad_norm": 2.3613460063934326,
842
+ "learning_rate": 3.0771428571428573e-06,
843
+ "loss": 0.4696,
844
+ "step": 2925
845
+ },
846
+ {
847
+ "epoch": 21.071428571428573,
848
+ "grad_norm": 1.5449374914169312,
849
+ "learning_rate": 3.005714285714286e-06,
850
+ "loss": 0.4753,
851
+ "step": 2950
852
+ },
853
+ {
854
+ "epoch": 21.25,
855
+ "grad_norm": 1.9528214931488037,
856
+ "learning_rate": 2.9342857142857144e-06,
857
+ "loss": 0.4777,
858
+ "step": 2975
859
+ },
860
+ {
861
+ "epoch": 21.428571428571427,
862
+ "grad_norm": 1.8519726991653442,
863
+ "learning_rate": 2.8628571428571435e-06,
864
+ "loss": 0.4739,
865
+ "step": 3000
866
+ },
867
+ {
868
+ "epoch": 21.428571428571427,
869
+ "eval_loss": 0.44868698716163635,
870
+ "eval_runtime": 10.7576,
871
+ "eval_samples_per_second": 46.2,
872
+ "eval_steps_per_second": 5.856,
873
+ "step": 3000
874
+ }
875
+ ],
876
+ "logging_steps": 25,
877
+ "max_steps": 4000,
878
+ "num_input_tokens_seen": 0,
879
+ "num_train_epochs": 29,
880
+ "save_steps": 1000,
881
+ "stateful_callbacks": {
882
+ "TrainerControl": {
883
+ "args": {
884
+ "should_epoch_stop": false,
885
+ "should_evaluate": false,
886
+ "should_log": false,
887
+ "should_save": true,
888
+ "should_training_stop": false
889
+ },
890
+ "attributes": {}
891
+ }
892
+ },
893
+ "total_flos": 8896616922279240.0,
894
+ "train_batch_size": 16,
895
+ "trial_name": null,
896
+ "trial_params": null
897
+ }
checkpoint-3000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a01e8c0a090b02699d5b506a30ffaf176234c63870a5a741f73db593b385337
3
+ size 5560
checkpoint-4000/added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "<ctc_blank>": 80,
3
+ "<mask>": 79
4
+ }
checkpoint-4000/config.json ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/speecht5_tts",
3
+ "activation_dropout": 0.1,
4
+ "apply_spec_augment": true,
5
+ "architectures": [
6
+ "SpeechT5ForTextToSpeech"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 0,
10
+ "conv_bias": false,
11
+ "conv_dim": [
12
+ 512,
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512
19
+ ],
20
+ "conv_kernel": [
21
+ 10,
22
+ 3,
23
+ 3,
24
+ 3,
25
+ 3,
26
+ 2,
27
+ 2
28
+ ],
29
+ "conv_stride": [
30
+ 5,
31
+ 2,
32
+ 2,
33
+ 2,
34
+ 2,
35
+ 2,
36
+ 2
37
+ ],
38
+ "decoder_attention_heads": 12,
39
+ "decoder_ffn_dim": 3072,
40
+ "decoder_layerdrop": 0.1,
41
+ "decoder_layers": 6,
42
+ "decoder_start_token_id": 2,
43
+ "encoder_attention_heads": 12,
44
+ "encoder_ffn_dim": 3072,
45
+ "encoder_layerdrop": 0.1,
46
+ "encoder_layers": 12,
47
+ "encoder_max_relative_position": 160,
48
+ "eos_token_id": 2,
49
+ "feat_extract_activation": "gelu",
50
+ "feat_extract_norm": "group",
51
+ "feat_proj_dropout": 0.0,
52
+ "guided_attention_loss_num_heads": 2,
53
+ "guided_attention_loss_scale": 10.0,
54
+ "guided_attention_loss_sigma": 0.4,
55
+ "hidden_act": "gelu",
56
+ "hidden_dropout": 0.1,
57
+ "hidden_size": 768,
58
+ "initializer_range": 0.02,
59
+ "is_encoder_decoder": true,
60
+ "layer_norm_eps": 1e-05,
61
+ "mask_feature_length": 10,
62
+ "mask_feature_min_masks": 0,
63
+ "mask_feature_prob": 0.0,
64
+ "mask_time_length": 10,
65
+ "mask_time_min_masks": 2,
66
+ "mask_time_prob": 0.05,
67
+ "max_length": 1876,
68
+ "max_speech_positions": 1876,
69
+ "max_text_positions": 600,
70
+ "model_type": "speecht5",
71
+ "num_conv_pos_embedding_groups": 16,
72
+ "num_conv_pos_embeddings": 128,
73
+ "num_feat_extract_layers": 7,
74
+ "num_mel_bins": 80,
75
+ "pad_token_id": 1,
76
+ "positional_dropout": 0.1,
77
+ "reduction_factor": 2,
78
+ "scale_embedding": false,
79
+ "speaker_embedding_dim": 512,
80
+ "speech_decoder_postnet_dropout": 0.5,
81
+ "speech_decoder_postnet_kernel": 5,
82
+ "speech_decoder_postnet_layers": 5,
83
+ "speech_decoder_postnet_units": 256,
84
+ "speech_decoder_prenet_dropout": 0.5,
85
+ "speech_decoder_prenet_layers": 2,
86
+ "speech_decoder_prenet_units": 256,
87
+ "torch_dtype": "float32",
88
+ "transformers_version": "4.50.0.dev0",
89
+ "use_cache": false,
90
+ "use_guided_attention_loss": true,
91
+ "vocab_size": 81
92
+ }
checkpoint-4000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a7f6c078a89790d6ee06ed66975fbf5180d360efd6cd656173ba550e5c4ba085
3
+ size 577789320
checkpoint-4000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2eb1e8a9ed6ee3d9b101138a9aeb165d43e422a016f1ef54cedf8a7c4fdf8c6d
3
+ size 1155772233
checkpoint-4000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fa285bbb0d5868c090a0912d03b0c8f9b6ffbfc0b058d94be0776891decf5ab7
3
+ size 14244
checkpoint-4000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:05ebaa20c6824a9fdb798c4b9ced025fecb94428858b5fc9ab59b75a52f7019a
3
+ size 988
checkpoint-4000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4783d1aad67332b882158dbf71f702c2c19d09ed6c5f6dbf2a69dcc433c36f68
3
+ size 1064
checkpoint-4000/special_tokens_map.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "eos_token": "</s>",
4
+ "mask_token": {
5
+ "content": "<mask>",
6
+ "lstrip": true,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false
10
+ },
11
+ "pad_token": "<pad>",
12
+ "unk_token": "<unk>"
13
+ }
checkpoint-4000/spm_char.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7fcc48f3e225f627b1641db410ceb0c8649bd2b0c982e150b03f8be3728ab560
3
+ size 238473
checkpoint-4000/tokenizer_config.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "79": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": true,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "80": {
44
+ "content": "<ctc_blank>",
45
+ "lstrip": false,
46
+ "normalized": true,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": false
50
+ }
51
+ },
52
+ "bos_token": "<s>",
53
+ "clean_up_tokenization_spaces": false,
54
+ "eos_token": "</s>",
55
+ "extra_special_tokens": {},
56
+ "mask_token": "<mask>",
57
+ "model_max_length": 600,
58
+ "normalize": false,
59
+ "pad_token": "<pad>",
60
+ "processor_class": "SpeechT5Processor",
61
+ "sp_model_kwargs": {},
62
+ "tokenizer_class": "SpeechT5Tokenizer",
63
+ "unk_token": "<unk>"
64
+ }
checkpoint-4000/trainer_state.json ADDED
@@ -0,0 +1,1185 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.4465157091617584,
3
+ "best_model_checkpoint": "./speecht5_tts_common_voice_17_0_id/checkpoint-4000",
4
+ "epoch": 28.571428571428573,
5
+ "eval_steps": 1000,
6
+ "global_step": 4000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.17857142857142858,
13
+ "grad_norm": 4.796061038970947,
14
+ "learning_rate": 4.6000000000000004e-07,
15
+ "loss": 1.0232,
16
+ "step": 25
17
+ },
18
+ {
19
+ "epoch": 0.35714285714285715,
20
+ "grad_norm": 6.88640022277832,
21
+ "learning_rate": 9.600000000000001e-07,
22
+ "loss": 0.9412,
23
+ "step": 50
24
+ },
25
+ {
26
+ "epoch": 0.5357142857142857,
27
+ "grad_norm": 4.282207489013672,
28
+ "learning_rate": 1.46e-06,
29
+ "loss": 0.8431,
30
+ "step": 75
31
+ },
32
+ {
33
+ "epoch": 0.7142857142857143,
34
+ "grad_norm": 4.095445156097412,
35
+ "learning_rate": 1.9600000000000003e-06,
36
+ "loss": 0.7959,
37
+ "step": 100
38
+ },
39
+ {
40
+ "epoch": 0.8928571428571429,
41
+ "grad_norm": 6.723930835723877,
42
+ "learning_rate": 2.46e-06,
43
+ "loss": 0.7825,
44
+ "step": 125
45
+ },
46
+ {
47
+ "epoch": 1.0714285714285714,
48
+ "grad_norm": 2.318833589553833,
49
+ "learning_rate": 2.96e-06,
50
+ "loss": 0.7967,
51
+ "step": 150
52
+ },
53
+ {
54
+ "epoch": 1.25,
55
+ "grad_norm": 2.1883444786071777,
56
+ "learning_rate": 3.46e-06,
57
+ "loss": 0.7826,
58
+ "step": 175
59
+ },
60
+ {
61
+ "epoch": 1.4285714285714286,
62
+ "grad_norm": 2.7793357372283936,
63
+ "learning_rate": 3.96e-06,
64
+ "loss": 0.7238,
65
+ "step": 200
66
+ },
67
+ {
68
+ "epoch": 1.6071428571428572,
69
+ "grad_norm": 3.0292282104492188,
70
+ "learning_rate": 4.4600000000000005e-06,
71
+ "loss": 0.6981,
72
+ "step": 225
73
+ },
74
+ {
75
+ "epoch": 1.7857142857142856,
76
+ "grad_norm": 4.8227314949035645,
77
+ "learning_rate": 4.960000000000001e-06,
78
+ "loss": 0.6753,
79
+ "step": 250
80
+ },
81
+ {
82
+ "epoch": 1.9642857142857144,
83
+ "grad_norm": 2.781235694885254,
84
+ "learning_rate": 5.460000000000001e-06,
85
+ "loss": 0.6337,
86
+ "step": 275
87
+ },
88
+ {
89
+ "epoch": 2.142857142857143,
90
+ "grad_norm": 2.9240171909332275,
91
+ "learning_rate": 5.9600000000000005e-06,
92
+ "loss": 0.6499,
93
+ "step": 300
94
+ },
95
+ {
96
+ "epoch": 2.3214285714285716,
97
+ "grad_norm": 2.0720555782318115,
98
+ "learning_rate": 6.460000000000001e-06,
99
+ "loss": 0.5995,
100
+ "step": 325
101
+ },
102
+ {
103
+ "epoch": 2.5,
104
+ "grad_norm": 3.025918483734131,
105
+ "learning_rate": 6.96e-06,
106
+ "loss": 0.6054,
107
+ "step": 350
108
+ },
109
+ {
110
+ "epoch": 2.678571428571429,
111
+ "grad_norm": 3.3047034740448,
112
+ "learning_rate": 7.4600000000000006e-06,
113
+ "loss": 0.5828,
114
+ "step": 375
115
+ },
116
+ {
117
+ "epoch": 2.857142857142857,
118
+ "grad_norm": 2.4087064266204834,
119
+ "learning_rate": 7.960000000000002e-06,
120
+ "loss": 0.5706,
121
+ "step": 400
122
+ },
123
+ {
124
+ "epoch": 3.0357142857142856,
125
+ "grad_norm": 2.7837984561920166,
126
+ "learning_rate": 8.46e-06,
127
+ "loss": 0.579,
128
+ "step": 425
129
+ },
130
+ {
131
+ "epoch": 3.2142857142857144,
132
+ "grad_norm": 1.7530593872070312,
133
+ "learning_rate": 8.96e-06,
134
+ "loss": 0.5455,
135
+ "step": 450
136
+ },
137
+ {
138
+ "epoch": 3.392857142857143,
139
+ "grad_norm": 2.3677515983581543,
140
+ "learning_rate": 9.460000000000001e-06,
141
+ "loss": 0.5557,
142
+ "step": 475
143
+ },
144
+ {
145
+ "epoch": 3.571428571428571,
146
+ "grad_norm": 2.445343494415283,
147
+ "learning_rate": 9.960000000000001e-06,
148
+ "loss": 0.5686,
149
+ "step": 500
150
+ },
151
+ {
152
+ "epoch": 3.75,
153
+ "grad_norm": 4.732877731323242,
154
+ "learning_rate": 9.934285714285715e-06,
155
+ "loss": 0.5635,
156
+ "step": 525
157
+ },
158
+ {
159
+ "epoch": 3.928571428571429,
160
+ "grad_norm": 4.45880126953125,
161
+ "learning_rate": 9.862857142857144e-06,
162
+ "loss": 0.5477,
163
+ "step": 550
164
+ },
165
+ {
166
+ "epoch": 4.107142857142857,
167
+ "grad_norm": 3.9657559394836426,
168
+ "learning_rate": 9.791428571428571e-06,
169
+ "loss": 0.5563,
170
+ "step": 575
171
+ },
172
+ {
173
+ "epoch": 4.285714285714286,
174
+ "grad_norm": 2.676309823989868,
175
+ "learning_rate": 9.72e-06,
176
+ "loss": 0.5221,
177
+ "step": 600
178
+ },
179
+ {
180
+ "epoch": 4.464285714285714,
181
+ "grad_norm": 2.89937686920166,
182
+ "learning_rate": 9.648571428571429e-06,
183
+ "loss": 0.5211,
184
+ "step": 625
185
+ },
186
+ {
187
+ "epoch": 4.642857142857143,
188
+ "grad_norm": 2.9824204444885254,
189
+ "learning_rate": 9.577142857142858e-06,
190
+ "loss": 0.5583,
191
+ "step": 650
192
+ },
193
+ {
194
+ "epoch": 4.821428571428571,
195
+ "grad_norm": 3.632369041442871,
196
+ "learning_rate": 9.505714285714287e-06,
197
+ "loss": 0.526,
198
+ "step": 675
199
+ },
200
+ {
201
+ "epoch": 5.0,
202
+ "grad_norm": 2.352837324142456,
203
+ "learning_rate": 9.434285714285714e-06,
204
+ "loss": 0.5211,
205
+ "step": 700
206
+ },
207
+ {
208
+ "epoch": 5.178571428571429,
209
+ "grad_norm": 1.3236348628997803,
210
+ "learning_rate": 9.362857142857143e-06,
211
+ "loss": 0.5173,
212
+ "step": 725
213
+ },
214
+ {
215
+ "epoch": 5.357142857142857,
216
+ "grad_norm": 1.8493444919586182,
217
+ "learning_rate": 9.291428571428572e-06,
218
+ "loss": 0.5094,
219
+ "step": 750
220
+ },
221
+ {
222
+ "epoch": 5.535714285714286,
223
+ "grad_norm": 1.7403544187545776,
224
+ "learning_rate": 9.220000000000002e-06,
225
+ "loss": 0.5239,
226
+ "step": 775
227
+ },
228
+ {
229
+ "epoch": 5.714285714285714,
230
+ "grad_norm": 4.10603666305542,
231
+ "learning_rate": 9.148571428571429e-06,
232
+ "loss": 0.5304,
233
+ "step": 800
234
+ },
235
+ {
236
+ "epoch": 5.892857142857143,
237
+ "grad_norm": 2.69100284576416,
238
+ "learning_rate": 9.077142857142858e-06,
239
+ "loss": 0.5224,
240
+ "step": 825
241
+ },
242
+ {
243
+ "epoch": 6.071428571428571,
244
+ "grad_norm": 1.956499695777893,
245
+ "learning_rate": 9.005714285714287e-06,
246
+ "loss": 0.5065,
247
+ "step": 850
248
+ },
249
+ {
250
+ "epoch": 6.25,
251
+ "grad_norm": 1.915540337562561,
252
+ "learning_rate": 8.934285714285716e-06,
253
+ "loss": 0.511,
254
+ "step": 875
255
+ },
256
+ {
257
+ "epoch": 6.428571428571429,
258
+ "grad_norm": 2.1903812885284424,
259
+ "learning_rate": 8.862857142857143e-06,
260
+ "loss": 0.5132,
261
+ "step": 900
262
+ },
263
+ {
264
+ "epoch": 6.607142857142857,
265
+ "grad_norm": 1.666169285774231,
266
+ "learning_rate": 8.791428571428572e-06,
267
+ "loss": 0.5362,
268
+ "step": 925
269
+ },
270
+ {
271
+ "epoch": 6.785714285714286,
272
+ "grad_norm": 1.6485190391540527,
273
+ "learning_rate": 8.720000000000001e-06,
274
+ "loss": 0.5115,
275
+ "step": 950
276
+ },
277
+ {
278
+ "epoch": 6.964285714285714,
279
+ "grad_norm": 1.7726221084594727,
280
+ "learning_rate": 8.64857142857143e-06,
281
+ "loss": 0.5183,
282
+ "step": 975
283
+ },
284
+ {
285
+ "epoch": 7.142857142857143,
286
+ "grad_norm": 2.7793469429016113,
287
+ "learning_rate": 8.577142857142858e-06,
288
+ "loss": 0.5133,
289
+ "step": 1000
290
+ },
291
+ {
292
+ "epoch": 7.142857142857143,
293
+ "eval_loss": 0.4698790907859802,
294
+ "eval_runtime": 10.706,
295
+ "eval_samples_per_second": 46.422,
296
+ "eval_steps_per_second": 5.885,
297
+ "step": 1000
298
+ },
299
+ {
300
+ "epoch": 7.321428571428571,
301
+ "grad_norm": 2.8019936084747314,
302
+ "learning_rate": 8.505714285714287e-06,
303
+ "loss": 0.518,
304
+ "step": 1025
305
+ },
306
+ {
307
+ "epoch": 7.5,
308
+ "grad_norm": 2.1961379051208496,
309
+ "learning_rate": 8.434285714285716e-06,
310
+ "loss": 0.5112,
311
+ "step": 1050
312
+ },
313
+ {
314
+ "epoch": 7.678571428571429,
315
+ "grad_norm": 2.5158090591430664,
316
+ "learning_rate": 8.362857142857143e-06,
317
+ "loss": 0.5091,
318
+ "step": 1075
319
+ },
320
+ {
321
+ "epoch": 7.857142857142857,
322
+ "grad_norm": 2.2158660888671875,
323
+ "learning_rate": 8.291428571428572e-06,
324
+ "loss": 0.5121,
325
+ "step": 1100
326
+ },
327
+ {
328
+ "epoch": 8.035714285714286,
329
+ "grad_norm": 2.8917131423950195,
330
+ "learning_rate": 8.220000000000001e-06,
331
+ "loss": 0.5134,
332
+ "step": 1125
333
+ },
334
+ {
335
+ "epoch": 8.214285714285714,
336
+ "grad_norm": 1.9233217239379883,
337
+ "learning_rate": 8.148571428571428e-06,
338
+ "loss": 0.5101,
339
+ "step": 1150
340
+ },
341
+ {
342
+ "epoch": 8.392857142857142,
343
+ "grad_norm": 2.437870979309082,
344
+ "learning_rate": 8.077142857142857e-06,
345
+ "loss": 0.5138,
346
+ "step": 1175
347
+ },
348
+ {
349
+ "epoch": 8.571428571428571,
350
+ "grad_norm": 1.7921634912490845,
351
+ "learning_rate": 8.005714285714286e-06,
352
+ "loss": 0.5155,
353
+ "step": 1200
354
+ },
355
+ {
356
+ "epoch": 8.75,
357
+ "grad_norm": 3.1069226264953613,
358
+ "learning_rate": 7.934285714285715e-06,
359
+ "loss": 0.5144,
360
+ "step": 1225
361
+ },
362
+ {
363
+ "epoch": 8.928571428571429,
364
+ "grad_norm": 1.7902675867080688,
365
+ "learning_rate": 7.862857142857143e-06,
366
+ "loss": 0.4973,
367
+ "step": 1250
368
+ },
369
+ {
370
+ "epoch": 9.107142857142858,
371
+ "grad_norm": 2.353273391723633,
372
+ "learning_rate": 7.791428571428572e-06,
373
+ "loss": 0.5035,
374
+ "step": 1275
375
+ },
376
+ {
377
+ "epoch": 9.285714285714286,
378
+ "grad_norm": 2.3287203311920166,
379
+ "learning_rate": 7.72e-06,
380
+ "loss": 0.5085,
381
+ "step": 1300
382
+ },
383
+ {
384
+ "epoch": 9.464285714285714,
385
+ "grad_norm": 1.6508110761642456,
386
+ "learning_rate": 7.64857142857143e-06,
387
+ "loss": 0.5116,
388
+ "step": 1325
389
+ },
390
+ {
391
+ "epoch": 9.642857142857142,
392
+ "grad_norm": 1.8232289552688599,
393
+ "learning_rate": 7.577142857142857e-06,
394
+ "loss": 0.5033,
395
+ "step": 1350
396
+ },
397
+ {
398
+ "epoch": 9.821428571428571,
399
+ "grad_norm": 2.0248517990112305,
400
+ "learning_rate": 7.505714285714286e-06,
401
+ "loss": 0.4974,
402
+ "step": 1375
403
+ },
404
+ {
405
+ "epoch": 10.0,
406
+ "grad_norm": 3.7484123706817627,
407
+ "learning_rate": 7.434285714285715e-06,
408
+ "loss": 0.4947,
409
+ "step": 1400
410
+ },
411
+ {
412
+ "epoch": 10.178571428571429,
413
+ "grad_norm": 1.793626308441162,
414
+ "learning_rate": 7.362857142857144e-06,
415
+ "loss": 0.5008,
416
+ "step": 1425
417
+ },
418
+ {
419
+ "epoch": 10.357142857142858,
420
+ "grad_norm": 1.8825795650482178,
421
+ "learning_rate": 7.291428571428571e-06,
422
+ "loss": 0.4965,
423
+ "step": 1450
424
+ },
425
+ {
426
+ "epoch": 10.535714285714286,
427
+ "grad_norm": 2.0584166049957275,
428
+ "learning_rate": 7.22e-06,
429
+ "loss": 0.4946,
430
+ "step": 1475
431
+ },
432
+ {
433
+ "epoch": 10.714285714285714,
434
+ "grad_norm": 3.3335680961608887,
435
+ "learning_rate": 7.148571428571429e-06,
436
+ "loss": 0.5048,
437
+ "step": 1500
438
+ },
439
+ {
440
+ "epoch": 10.892857142857142,
441
+ "grad_norm": 1.9972820281982422,
442
+ "learning_rate": 7.077142857142858e-06,
443
+ "loss": 0.5071,
444
+ "step": 1525
445
+ },
446
+ {
447
+ "epoch": 11.071428571428571,
448
+ "grad_norm": 2.2279226779937744,
449
+ "learning_rate": 7.0057142857142865e-06,
450
+ "loss": 0.4874,
451
+ "step": 1550
452
+ },
453
+ {
454
+ "epoch": 11.25,
455
+ "grad_norm": 2.168933868408203,
456
+ "learning_rate": 6.934285714285715e-06,
457
+ "loss": 0.5011,
458
+ "step": 1575
459
+ },
460
+ {
461
+ "epoch": 11.428571428571429,
462
+ "grad_norm": 3.186655044555664,
463
+ "learning_rate": 6.862857142857144e-06,
464
+ "loss": 0.4963,
465
+ "step": 1600
466
+ },
467
+ {
468
+ "epoch": 11.607142857142858,
469
+ "grad_norm": 2.6348812580108643,
470
+ "learning_rate": 6.791428571428572e-06,
471
+ "loss": 0.5007,
472
+ "step": 1625
473
+ },
474
+ {
475
+ "epoch": 11.785714285714286,
476
+ "grad_norm": 3.1159684658050537,
477
+ "learning_rate": 6.720000000000001e-06,
478
+ "loss": 0.4997,
479
+ "step": 1650
480
+ },
481
+ {
482
+ "epoch": 11.964285714285714,
483
+ "grad_norm": 4.068305492401123,
484
+ "learning_rate": 6.648571428571429e-06,
485
+ "loss": 0.4919,
486
+ "step": 1675
487
+ },
488
+ {
489
+ "epoch": 12.142857142857142,
490
+ "grad_norm": 2.315338134765625,
491
+ "learning_rate": 6.577142857142857e-06,
492
+ "loss": 0.4937,
493
+ "step": 1700
494
+ },
495
+ {
496
+ "epoch": 12.321428571428571,
497
+ "grad_norm": 1.5482977628707886,
498
+ "learning_rate": 6.505714285714286e-06,
499
+ "loss": 0.4964,
500
+ "step": 1725
501
+ },
502
+ {
503
+ "epoch": 12.5,
504
+ "grad_norm": 2.1527745723724365,
505
+ "learning_rate": 6.434285714285715e-06,
506
+ "loss": 0.494,
507
+ "step": 1750
508
+ },
509
+ {
510
+ "epoch": 12.678571428571429,
511
+ "grad_norm": 2.216703176498413,
512
+ "learning_rate": 6.3628571428571426e-06,
513
+ "loss": 0.4853,
514
+ "step": 1775
515
+ },
516
+ {
517
+ "epoch": 12.857142857142858,
518
+ "grad_norm": 2.939267873764038,
519
+ "learning_rate": 6.2914285714285716e-06,
520
+ "loss": 0.4978,
521
+ "step": 1800
522
+ },
523
+ {
524
+ "epoch": 13.035714285714286,
525
+ "grad_norm": 2.3067381381988525,
526
+ "learning_rate": 6.220000000000001e-06,
527
+ "loss": 0.5023,
528
+ "step": 1825
529
+ },
530
+ {
531
+ "epoch": 13.214285714285714,
532
+ "grad_norm": 2.2849552631378174,
533
+ "learning_rate": 6.14857142857143e-06,
534
+ "loss": 0.5013,
535
+ "step": 1850
536
+ },
537
+ {
538
+ "epoch": 13.392857142857142,
539
+ "grad_norm": 1.8568611145019531,
540
+ "learning_rate": 6.077142857142858e-06,
541
+ "loss": 0.4855,
542
+ "step": 1875
543
+ },
544
+ {
545
+ "epoch": 13.571428571428571,
546
+ "grad_norm": 2.784388542175293,
547
+ "learning_rate": 6.005714285714286e-06,
548
+ "loss": 0.4832,
549
+ "step": 1900
550
+ },
551
+ {
552
+ "epoch": 13.75,
553
+ "grad_norm": 1.718732237815857,
554
+ "learning_rate": 5.934285714285715e-06,
555
+ "loss": 0.4926,
556
+ "step": 1925
557
+ },
558
+ {
559
+ "epoch": 13.928571428571429,
560
+ "grad_norm": 2.3010714054107666,
561
+ "learning_rate": 5.862857142857143e-06,
562
+ "loss": 0.494,
563
+ "step": 1950
564
+ },
565
+ {
566
+ "epoch": 14.107142857142858,
567
+ "grad_norm": 2.0058493614196777,
568
+ "learning_rate": 5.791428571428572e-06,
569
+ "loss": 0.4901,
570
+ "step": 1975
571
+ },
572
+ {
573
+ "epoch": 14.285714285714286,
574
+ "grad_norm": 2.311893939971924,
575
+ "learning_rate": 5.72e-06,
576
+ "loss": 0.4774,
577
+ "step": 2000
578
+ },
579
+ {
580
+ "epoch": 14.285714285714286,
581
+ "eval_loss": 0.45390585064888,
582
+ "eval_runtime": 18.3883,
583
+ "eval_samples_per_second": 27.028,
584
+ "eval_steps_per_second": 3.426,
585
+ "step": 2000
586
+ },
587
+ {
588
+ "epoch": 14.464285714285714,
589
+ "grad_norm": 5.475688457489014,
590
+ "learning_rate": 5.6485714285714285e-06,
591
+ "loss": 0.479,
592
+ "step": 2025
593
+ },
594
+ {
595
+ "epoch": 14.642857142857142,
596
+ "grad_norm": 2.333171844482422,
597
+ "learning_rate": 5.5771428571428575e-06,
598
+ "loss": 0.487,
599
+ "step": 2050
600
+ },
601
+ {
602
+ "epoch": 14.821428571428571,
603
+ "grad_norm": 1.7914478778839111,
604
+ "learning_rate": 5.5057142857142865e-06,
605
+ "loss": 0.5042,
606
+ "step": 2075
607
+ },
608
+ {
609
+ "epoch": 15.0,
610
+ "grad_norm": 2.343949317932129,
611
+ "learning_rate": 5.4342857142857155e-06,
612
+ "loss": 0.4994,
613
+ "step": 2100
614
+ },
615
+ {
616
+ "epoch": 15.178571428571429,
617
+ "grad_norm": 2.7536513805389404,
618
+ "learning_rate": 5.362857142857143e-06,
619
+ "loss": 0.492,
620
+ "step": 2125
621
+ },
622
+ {
623
+ "epoch": 15.357142857142858,
624
+ "grad_norm": 3.5288944244384766,
625
+ "learning_rate": 5.291428571428572e-06,
626
+ "loss": 0.4878,
627
+ "step": 2150
628
+ },
629
+ {
630
+ "epoch": 15.535714285714286,
631
+ "grad_norm": 4.280928611755371,
632
+ "learning_rate": 5.220000000000001e-06,
633
+ "loss": 0.4908,
634
+ "step": 2175
635
+ },
636
+ {
637
+ "epoch": 15.714285714285714,
638
+ "grad_norm": 3.1674723625183105,
639
+ "learning_rate": 5.14857142857143e-06,
640
+ "loss": 0.4896,
641
+ "step": 2200
642
+ },
643
+ {
644
+ "epoch": 15.892857142857142,
645
+ "grad_norm": 1.4766712188720703,
646
+ "learning_rate": 5.077142857142857e-06,
647
+ "loss": 0.4891,
648
+ "step": 2225
649
+ },
650
+ {
651
+ "epoch": 16.071428571428573,
652
+ "grad_norm": 2.7163197994232178,
653
+ "learning_rate": 5.005714285714286e-06,
654
+ "loss": 0.4901,
655
+ "step": 2250
656
+ },
657
+ {
658
+ "epoch": 16.25,
659
+ "grad_norm": 2.2624833583831787,
660
+ "learning_rate": 4.934285714285715e-06,
661
+ "loss": 0.4841,
662
+ "step": 2275
663
+ },
664
+ {
665
+ "epoch": 16.428571428571427,
666
+ "grad_norm": 3.0063624382019043,
667
+ "learning_rate": 4.862857142857143e-06,
668
+ "loss": 0.4826,
669
+ "step": 2300
670
+ },
671
+ {
672
+ "epoch": 16.607142857142858,
673
+ "grad_norm": 1.5605120658874512,
674
+ "learning_rate": 4.7914285714285715e-06,
675
+ "loss": 0.4808,
676
+ "step": 2325
677
+ },
678
+ {
679
+ "epoch": 16.785714285714285,
680
+ "grad_norm": 1.8184423446655273,
681
+ "learning_rate": 4.7200000000000005e-06,
682
+ "loss": 0.4743,
683
+ "step": 2350
684
+ },
685
+ {
686
+ "epoch": 16.964285714285715,
687
+ "grad_norm": 1.490218162536621,
688
+ "learning_rate": 4.648571428571429e-06,
689
+ "loss": 0.4848,
690
+ "step": 2375
691
+ },
692
+ {
693
+ "epoch": 17.142857142857142,
694
+ "grad_norm": 2.210409641265869,
695
+ "learning_rate": 4.577142857142858e-06,
696
+ "loss": 0.4814,
697
+ "step": 2400
698
+ },
699
+ {
700
+ "epoch": 17.321428571428573,
701
+ "grad_norm": 1.8189564943313599,
702
+ "learning_rate": 4.505714285714286e-06,
703
+ "loss": 0.4749,
704
+ "step": 2425
705
+ },
706
+ {
707
+ "epoch": 17.5,
708
+ "grad_norm": 2.2493181228637695,
709
+ "learning_rate": 4.434285714285715e-06,
710
+ "loss": 0.4835,
711
+ "step": 2450
712
+ },
713
+ {
714
+ "epoch": 17.678571428571427,
715
+ "grad_norm": 2.454582452774048,
716
+ "learning_rate": 4.362857142857143e-06,
717
+ "loss": 0.4891,
718
+ "step": 2475
719
+ },
720
+ {
721
+ "epoch": 17.857142857142858,
722
+ "grad_norm": 2.0914249420166016,
723
+ "learning_rate": 4.291428571428572e-06,
724
+ "loss": 0.4774,
725
+ "step": 2500
726
+ },
727
+ {
728
+ "epoch": 18.035714285714285,
729
+ "grad_norm": 1.6150003671646118,
730
+ "learning_rate": 4.22e-06,
731
+ "loss": 0.4826,
732
+ "step": 2525
733
+ },
734
+ {
735
+ "epoch": 18.214285714285715,
736
+ "grad_norm": 2.94893217086792,
737
+ "learning_rate": 4.148571428571429e-06,
738
+ "loss": 0.4894,
739
+ "step": 2550
740
+ },
741
+ {
742
+ "epoch": 18.392857142857142,
743
+ "grad_norm": 1.913024663925171,
744
+ "learning_rate": 4.0771428571428574e-06,
745
+ "loss": 0.4784,
746
+ "step": 2575
747
+ },
748
+ {
749
+ "epoch": 18.571428571428573,
750
+ "grad_norm": 2.2716126441955566,
751
+ "learning_rate": 4.0057142857142864e-06,
752
+ "loss": 0.4709,
753
+ "step": 2600
754
+ },
755
+ {
756
+ "epoch": 18.75,
757
+ "grad_norm": 3.5856099128723145,
758
+ "learning_rate": 3.934285714285715e-06,
759
+ "loss": 0.4885,
760
+ "step": 2625
761
+ },
762
+ {
763
+ "epoch": 18.928571428571427,
764
+ "grad_norm": 4.390285015106201,
765
+ "learning_rate": 3.862857142857143e-06,
766
+ "loss": 0.4828,
767
+ "step": 2650
768
+ },
769
+ {
770
+ "epoch": 19.107142857142858,
771
+ "grad_norm": 3.690558910369873,
772
+ "learning_rate": 3.7914285714285722e-06,
773
+ "loss": 0.4862,
774
+ "step": 2675
775
+ },
776
+ {
777
+ "epoch": 19.285714285714285,
778
+ "grad_norm": 3.1892879009246826,
779
+ "learning_rate": 3.7200000000000004e-06,
780
+ "loss": 0.4798,
781
+ "step": 2700
782
+ },
783
+ {
784
+ "epoch": 19.464285714285715,
785
+ "grad_norm": 2.4511208534240723,
786
+ "learning_rate": 3.648571428571429e-06,
787
+ "loss": 0.4844,
788
+ "step": 2725
789
+ },
790
+ {
791
+ "epoch": 19.642857142857142,
792
+ "grad_norm": 1.642425537109375,
793
+ "learning_rate": 3.5771428571428576e-06,
794
+ "loss": 0.4845,
795
+ "step": 2750
796
+ },
797
+ {
798
+ "epoch": 19.821428571428573,
799
+ "grad_norm": 1.53347909450531,
800
+ "learning_rate": 3.505714285714286e-06,
801
+ "loss": 0.4821,
802
+ "step": 2775
803
+ },
804
+ {
805
+ "epoch": 20.0,
806
+ "grad_norm": 2.1212124824523926,
807
+ "learning_rate": 3.4342857142857143e-06,
808
+ "loss": 0.4756,
809
+ "step": 2800
810
+ },
811
+ {
812
+ "epoch": 20.178571428571427,
813
+ "grad_norm": 2.1033811569213867,
814
+ "learning_rate": 3.3628571428571433e-06,
815
+ "loss": 0.4729,
816
+ "step": 2825
817
+ },
818
+ {
819
+ "epoch": 20.357142857142858,
820
+ "grad_norm": 1.9785221815109253,
821
+ "learning_rate": 3.2914285714285715e-06,
822
+ "loss": 0.4783,
823
+ "step": 2850
824
+ },
825
+ {
826
+ "epoch": 20.535714285714285,
827
+ "grad_norm": 1.7057210206985474,
828
+ "learning_rate": 3.2200000000000005e-06,
829
+ "loss": 0.4728,
830
+ "step": 2875
831
+ },
832
+ {
833
+ "epoch": 20.714285714285715,
834
+ "grad_norm": 2.1187682151794434,
835
+ "learning_rate": 3.1485714285714287e-06,
836
+ "loss": 0.4765,
837
+ "step": 2900
838
+ },
839
+ {
840
+ "epoch": 20.892857142857142,
841
+ "grad_norm": 2.3613460063934326,
842
+ "learning_rate": 3.0771428571428573e-06,
843
+ "loss": 0.4696,
844
+ "step": 2925
845
+ },
846
+ {
847
+ "epoch": 21.071428571428573,
848
+ "grad_norm": 1.5449374914169312,
849
+ "learning_rate": 3.005714285714286e-06,
850
+ "loss": 0.4753,
851
+ "step": 2950
852
+ },
853
+ {
854
+ "epoch": 21.25,
855
+ "grad_norm": 1.9528214931488037,
856
+ "learning_rate": 2.9342857142857144e-06,
857
+ "loss": 0.4777,
858
+ "step": 2975
859
+ },
860
+ {
861
+ "epoch": 21.428571428571427,
862
+ "grad_norm": 1.8519726991653442,
863
+ "learning_rate": 2.8628571428571435e-06,
864
+ "loss": 0.4739,
865
+ "step": 3000
866
+ },
867
+ {
868
+ "epoch": 21.428571428571427,
869
+ "eval_loss": 0.44868698716163635,
870
+ "eval_runtime": 10.7576,
871
+ "eval_samples_per_second": 46.2,
872
+ "eval_steps_per_second": 5.856,
873
+ "step": 3000
874
+ },
875
+ {
876
+ "epoch": 21.607142857142858,
877
+ "grad_norm": 1.5077314376831055,
878
+ "learning_rate": 2.7914285714285716e-06,
879
+ "loss": 0.4723,
880
+ "step": 3025
881
+ },
882
+ {
883
+ "epoch": 21.785714285714285,
884
+ "grad_norm": 1.6296398639678955,
885
+ "learning_rate": 2.7200000000000002e-06,
886
+ "loss": 0.4775,
887
+ "step": 3050
888
+ },
889
+ {
890
+ "epoch": 21.964285714285715,
891
+ "grad_norm": 2.155275821685791,
892
+ "learning_rate": 2.648571428571429e-06,
893
+ "loss": 0.4777,
894
+ "step": 3075
895
+ },
896
+ {
897
+ "epoch": 22.142857142857142,
898
+ "grad_norm": 1.4826477766036987,
899
+ "learning_rate": 2.5771428571428574e-06,
900
+ "loss": 0.4778,
901
+ "step": 3100
902
+ },
903
+ {
904
+ "epoch": 22.321428571428573,
905
+ "grad_norm": 4.133993148803711,
906
+ "learning_rate": 2.5057142857142856e-06,
907
+ "loss": 0.4867,
908
+ "step": 3125
909
+ },
910
+ {
911
+ "epoch": 22.5,
912
+ "grad_norm": 1.8221015930175781,
913
+ "learning_rate": 2.4342857142857146e-06,
914
+ "loss": 0.4843,
915
+ "step": 3150
916
+ },
917
+ {
918
+ "epoch": 22.678571428571427,
919
+ "grad_norm": 2.3254904747009277,
920
+ "learning_rate": 2.362857142857143e-06,
921
+ "loss": 0.4816,
922
+ "step": 3175
923
+ },
924
+ {
925
+ "epoch": 22.857142857142858,
926
+ "grad_norm": 1.7816497087478638,
927
+ "learning_rate": 2.2914285714285718e-06,
928
+ "loss": 0.4698,
929
+ "step": 3200
930
+ },
931
+ {
932
+ "epoch": 23.035714285714285,
933
+ "grad_norm": 2.3238487243652344,
934
+ "learning_rate": 2.2200000000000003e-06,
935
+ "loss": 0.4951,
936
+ "step": 3225
937
+ },
938
+ {
939
+ "epoch": 23.214285714285715,
940
+ "grad_norm": 2.038130044937134,
941
+ "learning_rate": 2.148571428571429e-06,
942
+ "loss": 0.4853,
943
+ "step": 3250
944
+ },
945
+ {
946
+ "epoch": 23.392857142857142,
947
+ "grad_norm": 2.0773870944976807,
948
+ "learning_rate": 2.077142857142857e-06,
949
+ "loss": 0.4674,
950
+ "step": 3275
951
+ },
952
+ {
953
+ "epoch": 23.571428571428573,
954
+ "grad_norm": 1.8331812620162964,
955
+ "learning_rate": 2.0057142857142857e-06,
956
+ "loss": 0.4748,
957
+ "step": 3300
958
+ },
959
+ {
960
+ "epoch": 23.75,
961
+ "grad_norm": 2.034414291381836,
962
+ "learning_rate": 1.9342857142857143e-06,
963
+ "loss": 0.4761,
964
+ "step": 3325
965
+ },
966
+ {
967
+ "epoch": 23.928571428571427,
968
+ "grad_norm": 2.0720574855804443,
969
+ "learning_rate": 1.8628571428571429e-06,
970
+ "loss": 0.4714,
971
+ "step": 3350
972
+ },
973
+ {
974
+ "epoch": 24.107142857142858,
975
+ "grad_norm": 1.5996496677398682,
976
+ "learning_rate": 1.7914285714285715e-06,
977
+ "loss": 0.4741,
978
+ "step": 3375
979
+ },
980
+ {
981
+ "epoch": 24.285714285714285,
982
+ "grad_norm": 1.812302827835083,
983
+ "learning_rate": 1.72e-06,
984
+ "loss": 0.4761,
985
+ "step": 3400
986
+ },
987
+ {
988
+ "epoch": 24.464285714285715,
989
+ "grad_norm": 3.9967708587646484,
990
+ "learning_rate": 1.6485714285714289e-06,
991
+ "loss": 0.4659,
992
+ "step": 3425
993
+ },
994
+ {
995
+ "epoch": 24.642857142857142,
996
+ "grad_norm": 2.3776981830596924,
997
+ "learning_rate": 1.5771428571428574e-06,
998
+ "loss": 0.4826,
999
+ "step": 3450
1000
+ },
1001
+ {
1002
+ "epoch": 24.821428571428573,
1003
+ "grad_norm": 2.67183780670166,
1004
+ "learning_rate": 1.5057142857142858e-06,
1005
+ "loss": 0.4808,
1006
+ "step": 3475
1007
+ },
1008
+ {
1009
+ "epoch": 25.0,
1010
+ "grad_norm": 2.5007424354553223,
1011
+ "learning_rate": 1.4342857142857144e-06,
1012
+ "loss": 0.4744,
1013
+ "step": 3500
1014
+ },
1015
+ {
1016
+ "epoch": 25.178571428571427,
1017
+ "grad_norm": 3.58581280708313,
1018
+ "learning_rate": 1.362857142857143e-06,
1019
+ "loss": 0.4861,
1020
+ "step": 3525
1021
+ },
1022
+ {
1023
+ "epoch": 25.357142857142858,
1024
+ "grad_norm": 2.0921339988708496,
1025
+ "learning_rate": 1.2914285714285716e-06,
1026
+ "loss": 0.472,
1027
+ "step": 3550
1028
+ },
1029
+ {
1030
+ "epoch": 25.535714285714285,
1031
+ "grad_norm": 1.7422988414764404,
1032
+ "learning_rate": 1.2200000000000002e-06,
1033
+ "loss": 0.4747,
1034
+ "step": 3575
1035
+ },
1036
+ {
1037
+ "epoch": 25.714285714285715,
1038
+ "grad_norm": 2.5243210792541504,
1039
+ "learning_rate": 1.1485714285714286e-06,
1040
+ "loss": 0.4826,
1041
+ "step": 3600
1042
+ },
1043
+ {
1044
+ "epoch": 25.892857142857142,
1045
+ "grad_norm": 1.8407930135726929,
1046
+ "learning_rate": 1.0771428571428574e-06,
1047
+ "loss": 0.4757,
1048
+ "step": 3625
1049
+ },
1050
+ {
1051
+ "epoch": 26.071428571428573,
1052
+ "grad_norm": 2.1124722957611084,
1053
+ "learning_rate": 1.0057142857142857e-06,
1054
+ "loss": 0.4749,
1055
+ "step": 3650
1056
+ },
1057
+ {
1058
+ "epoch": 26.25,
1059
+ "grad_norm": 2.7014122009277344,
1060
+ "learning_rate": 9.342857142857144e-07,
1061
+ "loss": 0.4737,
1062
+ "step": 3675
1063
+ },
1064
+ {
1065
+ "epoch": 26.428571428571427,
1066
+ "grad_norm": 1.5146933794021606,
1067
+ "learning_rate": 8.628571428571429e-07,
1068
+ "loss": 0.4704,
1069
+ "step": 3700
1070
+ },
1071
+ {
1072
+ "epoch": 26.607142857142858,
1073
+ "grad_norm": 1.398466944694519,
1074
+ "learning_rate": 7.914285714285715e-07,
1075
+ "loss": 0.4732,
1076
+ "step": 3725
1077
+ },
1078
+ {
1079
+ "epoch": 26.785714285714285,
1080
+ "grad_norm": 1.5718879699707031,
1081
+ "learning_rate": 7.2e-07,
1082
+ "loss": 0.4715,
1083
+ "step": 3750
1084
+ },
1085
+ {
1086
+ "epoch": 26.964285714285715,
1087
+ "grad_norm": 2.2760274410247803,
1088
+ "learning_rate": 6.485714285714287e-07,
1089
+ "loss": 0.4803,
1090
+ "step": 3775
1091
+ },
1092
+ {
1093
+ "epoch": 27.142857142857142,
1094
+ "grad_norm": 1.6266584396362305,
1095
+ "learning_rate": 5.771428571428572e-07,
1096
+ "loss": 0.4738,
1097
+ "step": 3800
1098
+ },
1099
+ {
1100
+ "epoch": 27.321428571428573,
1101
+ "grad_norm": 1.7928359508514404,
1102
+ "learning_rate": 5.057142857142858e-07,
1103
+ "loss": 0.4699,
1104
+ "step": 3825
1105
+ },
1106
+ {
1107
+ "epoch": 27.5,
1108
+ "grad_norm": 2.1646013259887695,
1109
+ "learning_rate": 4.342857142857143e-07,
1110
+ "loss": 0.4752,
1111
+ "step": 3850
1112
+ },
1113
+ {
1114
+ "epoch": 27.678571428571427,
1115
+ "grad_norm": 2.122992515563965,
1116
+ "learning_rate": 3.6285714285714283e-07,
1117
+ "loss": 0.4791,
1118
+ "step": 3875
1119
+ },
1120
+ {
1121
+ "epoch": 27.857142857142858,
1122
+ "grad_norm": 2.3135998249053955,
1123
+ "learning_rate": 2.914285714285715e-07,
1124
+ "loss": 0.4722,
1125
+ "step": 3900
1126
+ },
1127
+ {
1128
+ "epoch": 28.035714285714285,
1129
+ "grad_norm": 1.7877129316329956,
1130
+ "learning_rate": 2.2e-07,
1131
+ "loss": 0.4617,
1132
+ "step": 3925
1133
+ },
1134
+ {
1135
+ "epoch": 28.214285714285715,
1136
+ "grad_norm": 2.576960802078247,
1137
+ "learning_rate": 1.4857142857142857e-07,
1138
+ "loss": 0.4693,
1139
+ "step": 3950
1140
+ },
1141
+ {
1142
+ "epoch": 28.392857142857142,
1143
+ "grad_norm": 2.0099828243255615,
1144
+ "learning_rate": 7.714285714285715e-08,
1145
+ "loss": 0.4713,
1146
+ "step": 3975
1147
+ },
1148
+ {
1149
+ "epoch": 28.571428571428573,
1150
+ "grad_norm": 1.6247307062149048,
1151
+ "learning_rate": 5.714285714285715e-09,
1152
+ "loss": 0.4845,
1153
+ "step": 4000
1154
+ },
1155
+ {
1156
+ "epoch": 28.571428571428573,
1157
+ "eval_loss": 0.4465157091617584,
1158
+ "eval_runtime": 10.9109,
1159
+ "eval_samples_per_second": 45.551,
1160
+ "eval_steps_per_second": 5.774,
1161
+ "step": 4000
1162
+ }
1163
+ ],
1164
+ "logging_steps": 25,
1165
+ "max_steps": 4000,
1166
+ "num_input_tokens_seen": 0,
1167
+ "num_train_epochs": 29,
1168
+ "save_steps": 1000,
1169
+ "stateful_callbacks": {
1170
+ "TrainerControl": {
1171
+ "args": {
1172
+ "should_epoch_stop": false,
1173
+ "should_evaluate": false,
1174
+ "should_log": false,
1175
+ "should_save": true,
1176
+ "should_training_stop": true
1177
+ },
1178
+ "attributes": {}
1179
+ }
1180
+ },
1181
+ "total_flos": 1.186346947223592e+16,
1182
+ "train_batch_size": 16,
1183
+ "trial_name": null,
1184
+ "trial_params": null
1185
+ }
checkpoint-4000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a01e8c0a090b02699d5b506a30ffaf176234c63870a5a741f73db593b385337
3
+ size 5560