ai557 commited on
Commit
6e90a50
·
verified ·
1 Parent(s): 95a0471

Upload 13 files

Browse files
added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "<ctc_blank>": 80,
3
+ "<mask>": 79
4
+ }
config.json ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_dropout": 0.1,
3
+ "apply_spec_augment": true,
4
+ "architectures": [
5
+ "SpeechT5ForTextToSpeech"
6
+ ],
7
+ "attention_dropout": 0.1,
8
+ "bos_token_id": 0,
9
+ "conv_bias": false,
10
+ "conv_dim": [
11
+ 512,
12
+ 512,
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512
18
+ ],
19
+ "conv_kernel": [
20
+ 10,
21
+ 3,
22
+ 3,
23
+ 3,
24
+ 3,
25
+ 2,
26
+ 2
27
+ ],
28
+ "conv_stride": [
29
+ 5,
30
+ 2,
31
+ 2,
32
+ 2,
33
+ 2,
34
+ 2,
35
+ 2
36
+ ],
37
+ "decoder_attention_heads": 12,
38
+ "decoder_ffn_dim": 3072,
39
+ "decoder_layerdrop": 0.1,
40
+ "decoder_layers": 6,
41
+ "decoder_start_token_id": 2,
42
+ "encoder_attention_heads": 12,
43
+ "encoder_ffn_dim": 3072,
44
+ "encoder_layerdrop": 0.1,
45
+ "encoder_layers": 12,
46
+ "encoder_max_relative_position": 160,
47
+ "eos_token_id": 2,
48
+ "feat_extract_activation": "gelu",
49
+ "feat_extract_norm": "group",
50
+ "feat_proj_dropout": 0.0,
51
+ "guided_attention_loss_num_heads": 2,
52
+ "guided_attention_loss_scale": 10.0,
53
+ "guided_attention_loss_sigma": 0.4,
54
+ "hidden_act": "gelu",
55
+ "hidden_dropout": 0.1,
56
+ "hidden_size": 768,
57
+ "initializer_range": 0.02,
58
+ "is_encoder_decoder": true,
59
+ "layer_norm_eps": 1e-05,
60
+ "mask_feature_length": 10,
61
+ "mask_feature_min_masks": 0,
62
+ "mask_feature_prob": 0.0,
63
+ "mask_time_length": 10,
64
+ "mask_time_min_masks": 2,
65
+ "mask_time_prob": 0.05,
66
+ "max_length": null,
67
+ "max_speech_positions": 1876,
68
+ "max_text_positions": 600,
69
+ "model_type": "speecht5",
70
+ "num_conv_pos_embedding_groups": 16,
71
+ "num_conv_pos_embeddings": 128,
72
+ "num_feat_extract_layers": 7,
73
+ "num_mel_bins": 80,
74
+ "pad_token_id": 1,
75
+ "positional_dropout": 0.1,
76
+ "reduction_factor": 2,
77
+ "scale_embedding": false,
78
+ "speaker_embedding_dim": 512,
79
+ "speech_decoder_postnet_dropout": 0.5,
80
+ "speech_decoder_postnet_kernel": 5,
81
+ "speech_decoder_postnet_layers": 5,
82
+ "speech_decoder_postnet_units": 256,
83
+ "speech_decoder_prenet_dropout": 0.5,
84
+ "speech_decoder_prenet_layers": 2,
85
+ "speech_decoder_prenet_units": 256,
86
+ "torch_dtype": "float32",
87
+ "transformers_version": "4.53.1",
88
+ "use_cache": false,
89
+ "use_guided_attention_loss": true,
90
+ "vocab_size": 81
91
+ }
generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
+ "decoder_start_token_id": 2,
5
+ "eos_token_id": 2,
6
+ "max_length": 1876,
7
+ "pad_token_id": 1,
8
+ "transformers_version": "4.53.1"
9
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a1b2d6a76722f6816d4e4cc6de9aa7a6aee5fcaf72653e6a37f0b2d74be5f71c
3
+ size 577789320
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0d0b548d5a76319e114c6dd820d2ee597be0d730b315bcd898bf9434cfe6adc3
3
+ size 1155772634
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1317c85ea1e8711e98fa2b0eabe82c14f2d48c72144c2b2c6046bfe9f3dad7e1
3
+ size 14645
scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99b3a25ef62bcc235295487baccdd11a58ce1178762249833214489c9bfbef80
3
+ size 1383
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:29d64e8e67dc075981670f1e3676dffb33cd9f69d01f06875600c693b86402d6
3
+ size 1465
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
+ }
spm_char.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7fcc48f3e225f627b1641db410ceb0c8649bd2b0c982e150b03f8be3728ab560
3
+ size 238473
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
+ }
trainer_state.json ADDED
@@ -0,0 +1,626 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1000,
3
+ "best_metric": 0.7969154715538025,
4
+ "best_model_checkpoint": "speecht5_finetuned_nepali/checkpoint-1000",
5
+ "epoch": 2000.0,
6
+ "eval_steps": 500,
7
+ "global_step": 2000,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 25.0,
14
+ "grad_norm": 10.529245376586914,
15
+ "learning_rate": 9.5e-06,
16
+ "loss": 1.0362,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 50.0,
21
+ "grad_norm": 14.513840675354004,
22
+ "learning_rate": 2.2000000000000003e-05,
23
+ "loss": 0.4845,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 75.0,
28
+ "grad_norm": 22.360095977783203,
29
+ "learning_rate": 3.45e-05,
30
+ "loss": 0.415,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 100.0,
35
+ "grad_norm": 3.0262670516967773,
36
+ "learning_rate": 4.7e-05,
37
+ "loss": 0.2888,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 125.0,
42
+ "grad_norm": 3.495602607727051,
43
+ "learning_rate": 5.95e-05,
44
+ "loss": 0.3118,
45
+ "step": 125
46
+ },
47
+ {
48
+ "epoch": 150.0,
49
+ "grad_norm": 12.870991706848145,
50
+ "learning_rate": 7.2e-05,
51
+ "loss": 0.2475,
52
+ "step": 150
53
+ },
54
+ {
55
+ "epoch": 175.0,
56
+ "grad_norm": 25.45779800415039,
57
+ "learning_rate": 8.450000000000001e-05,
58
+ "loss": 0.2191,
59
+ "step": 175
60
+ },
61
+ {
62
+ "epoch": 200.0,
63
+ "grad_norm": 16.621234893798828,
64
+ "learning_rate": 9.7e-05,
65
+ "loss": 0.2357,
66
+ "step": 200
67
+ },
68
+ {
69
+ "epoch": 225.0,
70
+ "grad_norm": 3.157877206802368,
71
+ "learning_rate": 9.894444444444446e-05,
72
+ "loss": 0.2045,
73
+ "step": 225
74
+ },
75
+ {
76
+ "epoch": 250.0,
77
+ "grad_norm": 3.6257405281066895,
78
+ "learning_rate": 9.755555555555555e-05,
79
+ "loss": 0.2164,
80
+ "step": 250
81
+ },
82
+ {
83
+ "epoch": 275.0,
84
+ "grad_norm": 3.596989870071411,
85
+ "learning_rate": 9.616666666666667e-05,
86
+ "loss": 0.2195,
87
+ "step": 275
88
+ },
89
+ {
90
+ "epoch": 300.0,
91
+ "grad_norm": 9.64238166809082,
92
+ "learning_rate": 9.477777777777779e-05,
93
+ "loss": 0.2001,
94
+ "step": 300
95
+ },
96
+ {
97
+ "epoch": 325.0,
98
+ "grad_norm": 2.2049977779388428,
99
+ "learning_rate": 9.33888888888889e-05,
100
+ "loss": 0.2109,
101
+ "step": 325
102
+ },
103
+ {
104
+ "epoch": 350.0,
105
+ "grad_norm": 13.762038230895996,
106
+ "learning_rate": 9.200000000000001e-05,
107
+ "loss": 0.1858,
108
+ "step": 350
109
+ },
110
+ {
111
+ "epoch": 375.0,
112
+ "grad_norm": 16.161832809448242,
113
+ "learning_rate": 9.061111111111112e-05,
114
+ "loss": 0.2076,
115
+ "step": 375
116
+ },
117
+ {
118
+ "epoch": 400.0,
119
+ "grad_norm": 4.481337070465088,
120
+ "learning_rate": 8.922222222222223e-05,
121
+ "loss": 0.1771,
122
+ "step": 400
123
+ },
124
+ {
125
+ "epoch": 425.0,
126
+ "grad_norm": 4.335459232330322,
127
+ "learning_rate": 8.783333333333333e-05,
128
+ "loss": 0.1854,
129
+ "step": 425
130
+ },
131
+ {
132
+ "epoch": 450.0,
133
+ "grad_norm": 4.443661212921143,
134
+ "learning_rate": 8.644444444444445e-05,
135
+ "loss": 0.1635,
136
+ "step": 450
137
+ },
138
+ {
139
+ "epoch": 475.0,
140
+ "grad_norm": 5.2760796546936035,
141
+ "learning_rate": 8.505555555555556e-05,
142
+ "loss": 0.1767,
143
+ "step": 475
144
+ },
145
+ {
146
+ "epoch": 500.0,
147
+ "grad_norm": 13.874415397644043,
148
+ "learning_rate": 8.366666666666668e-05,
149
+ "loss": 0.1732,
150
+ "step": 500
151
+ },
152
+ {
153
+ "epoch": 500.0,
154
+ "eval_loss": 0.8349313735961914,
155
+ "eval_runtime": 0.0239,
156
+ "eval_samples_per_second": 41.882,
157
+ "eval_steps_per_second": 41.882,
158
+ "step": 500
159
+ },
160
+ {
161
+ "epoch": 525.0,
162
+ "grad_norm": 7.144843101501465,
163
+ "learning_rate": 8.227777777777778e-05,
164
+ "loss": 0.1622,
165
+ "step": 525
166
+ },
167
+ {
168
+ "epoch": 550.0,
169
+ "grad_norm": 9.63769245147705,
170
+ "learning_rate": 8.088888888888889e-05,
171
+ "loss": 0.1594,
172
+ "step": 550
173
+ },
174
+ {
175
+ "epoch": 575.0,
176
+ "grad_norm": 18.466543197631836,
177
+ "learning_rate": 7.950000000000001e-05,
178
+ "loss": 0.1584,
179
+ "step": 575
180
+ },
181
+ {
182
+ "epoch": 600.0,
183
+ "grad_norm": 4.807222843170166,
184
+ "learning_rate": 7.811111111111111e-05,
185
+ "loss": 0.1928,
186
+ "step": 600
187
+ },
188
+ {
189
+ "epoch": 625.0,
190
+ "grad_norm": 7.026573657989502,
191
+ "learning_rate": 7.672222222222223e-05,
192
+ "loss": 0.1403,
193
+ "step": 625
194
+ },
195
+ {
196
+ "epoch": 650.0,
197
+ "grad_norm": 1.3481919765472412,
198
+ "learning_rate": 7.533333333333334e-05,
199
+ "loss": 0.1454,
200
+ "step": 650
201
+ },
202
+ {
203
+ "epoch": 675.0,
204
+ "grad_norm": 1.3561595678329468,
205
+ "learning_rate": 7.394444444444445e-05,
206
+ "loss": 0.128,
207
+ "step": 675
208
+ },
209
+ {
210
+ "epoch": 700.0,
211
+ "grad_norm": 8.730093955993652,
212
+ "learning_rate": 7.255555555555555e-05,
213
+ "loss": 0.1306,
214
+ "step": 700
215
+ },
216
+ {
217
+ "epoch": 725.0,
218
+ "grad_norm": 1.6442005634307861,
219
+ "learning_rate": 7.116666666666667e-05,
220
+ "loss": 0.1332,
221
+ "step": 725
222
+ },
223
+ {
224
+ "epoch": 750.0,
225
+ "grad_norm": 2.5606157779693604,
226
+ "learning_rate": 6.977777777777779e-05,
227
+ "loss": 0.1289,
228
+ "step": 750
229
+ },
230
+ {
231
+ "epoch": 775.0,
232
+ "grad_norm": 10.112384796142578,
233
+ "learning_rate": 6.83888888888889e-05,
234
+ "loss": 0.1354,
235
+ "step": 775
236
+ },
237
+ {
238
+ "epoch": 800.0,
239
+ "grad_norm": 1.0964512825012207,
240
+ "learning_rate": 6.7e-05,
241
+ "loss": 0.1147,
242
+ "step": 800
243
+ },
244
+ {
245
+ "epoch": 825.0,
246
+ "grad_norm": 1.3629651069641113,
247
+ "learning_rate": 6.561111111111111e-05,
248
+ "loss": 0.1129,
249
+ "step": 825
250
+ },
251
+ {
252
+ "epoch": 850.0,
253
+ "grad_norm": 12.741216659545898,
254
+ "learning_rate": 6.422222222222223e-05,
255
+ "loss": 0.1098,
256
+ "step": 850
257
+ },
258
+ {
259
+ "epoch": 875.0,
260
+ "grad_norm": 2.6955878734588623,
261
+ "learning_rate": 6.283333333333333e-05,
262
+ "loss": 0.1103,
263
+ "step": 875
264
+ },
265
+ {
266
+ "epoch": 900.0,
267
+ "grad_norm": 1.8039289712905884,
268
+ "learning_rate": 6.144444444444445e-05,
269
+ "loss": 0.114,
270
+ "step": 900
271
+ },
272
+ {
273
+ "epoch": 925.0,
274
+ "grad_norm": 2.290956735610962,
275
+ "learning_rate": 6.005555555555555e-05,
276
+ "loss": 0.1182,
277
+ "step": 925
278
+ },
279
+ {
280
+ "epoch": 950.0,
281
+ "grad_norm": 0.9769994616508484,
282
+ "learning_rate": 5.866666666666667e-05,
283
+ "loss": 0.1236,
284
+ "step": 950
285
+ },
286
+ {
287
+ "epoch": 975.0,
288
+ "grad_norm": 0.7922130227088928,
289
+ "learning_rate": 5.7277777777777785e-05,
290
+ "loss": 0.1029,
291
+ "step": 975
292
+ },
293
+ {
294
+ "epoch": 1000.0,
295
+ "grad_norm": 0.9130666255950928,
296
+ "learning_rate": 5.588888888888889e-05,
297
+ "loss": 0.1032,
298
+ "step": 1000
299
+ },
300
+ {
301
+ "epoch": 1000.0,
302
+ "eval_loss": 0.7969154715538025,
303
+ "eval_runtime": 0.0205,
304
+ "eval_samples_per_second": 48.798,
305
+ "eval_steps_per_second": 48.798,
306
+ "step": 1000
307
+ },
308
+ {
309
+ "epoch": 1025.0,
310
+ "grad_norm": 3.981950283050537,
311
+ "learning_rate": 5.45e-05,
312
+ "loss": 0.1481,
313
+ "step": 1025
314
+ },
315
+ {
316
+ "epoch": 1050.0,
317
+ "grad_norm": 1.1825660467147827,
318
+ "learning_rate": 5.311111111111111e-05,
319
+ "loss": 0.104,
320
+ "step": 1050
321
+ },
322
+ {
323
+ "epoch": 1075.0,
324
+ "grad_norm": 2.0291762351989746,
325
+ "learning_rate": 5.172222222222223e-05,
326
+ "loss": 0.1049,
327
+ "step": 1075
328
+ },
329
+ {
330
+ "epoch": 1100.0,
331
+ "grad_norm": 0.8333075046539307,
332
+ "learning_rate": 5.0333333333333335e-05,
333
+ "loss": 0.0917,
334
+ "step": 1100
335
+ },
336
+ {
337
+ "epoch": 1125.0,
338
+ "grad_norm": 19.428544998168945,
339
+ "learning_rate": 4.894444444444445e-05,
340
+ "loss": 0.1106,
341
+ "step": 1125
342
+ },
343
+ {
344
+ "epoch": 1150.0,
345
+ "grad_norm": 1.1518399715423584,
346
+ "learning_rate": 4.755555555555556e-05,
347
+ "loss": 0.0911,
348
+ "step": 1150
349
+ },
350
+ {
351
+ "epoch": 1175.0,
352
+ "grad_norm": 11.079072952270508,
353
+ "learning_rate": 4.6166666666666666e-05,
354
+ "loss": 0.0853,
355
+ "step": 1175
356
+ },
357
+ {
358
+ "epoch": 1200.0,
359
+ "grad_norm": 3.140981435775757,
360
+ "learning_rate": 4.477777777777778e-05,
361
+ "loss": 0.1094,
362
+ "step": 1200
363
+ },
364
+ {
365
+ "epoch": 1225.0,
366
+ "grad_norm": 1.3652321100234985,
367
+ "learning_rate": 4.338888888888889e-05,
368
+ "loss": 0.085,
369
+ "step": 1225
370
+ },
371
+ {
372
+ "epoch": 1250.0,
373
+ "grad_norm": 4.102111339569092,
374
+ "learning_rate": 4.2e-05,
375
+ "loss": 0.0715,
376
+ "step": 1250
377
+ },
378
+ {
379
+ "epoch": 1275.0,
380
+ "grad_norm": 0.7971464395523071,
381
+ "learning_rate": 4.061111111111111e-05,
382
+ "loss": 0.0995,
383
+ "step": 1275
384
+ },
385
+ {
386
+ "epoch": 1300.0,
387
+ "grad_norm": 2.4831583499908447,
388
+ "learning_rate": 3.922222222222223e-05,
389
+ "loss": 0.0911,
390
+ "step": 1300
391
+ },
392
+ {
393
+ "epoch": 1325.0,
394
+ "grad_norm": 0.8036356568336487,
395
+ "learning_rate": 3.7833333333333336e-05,
396
+ "loss": 0.0744,
397
+ "step": 1325
398
+ },
399
+ {
400
+ "epoch": 1350.0,
401
+ "grad_norm": 0.7885045409202576,
402
+ "learning_rate": 3.644444444444445e-05,
403
+ "loss": 0.0819,
404
+ "step": 1350
405
+ },
406
+ {
407
+ "epoch": 1375.0,
408
+ "grad_norm": 0.8334118127822876,
409
+ "learning_rate": 3.505555555555556e-05,
410
+ "loss": 0.0814,
411
+ "step": 1375
412
+ },
413
+ {
414
+ "epoch": 1400.0,
415
+ "grad_norm": 3.9760286808013916,
416
+ "learning_rate": 3.366666666666667e-05,
417
+ "loss": 0.0787,
418
+ "step": 1400
419
+ },
420
+ {
421
+ "epoch": 1425.0,
422
+ "grad_norm": 1.063104271888733,
423
+ "learning_rate": 3.227777777777778e-05,
424
+ "loss": 0.0922,
425
+ "step": 1425
426
+ },
427
+ {
428
+ "epoch": 1450.0,
429
+ "grad_norm": 1.2303999662399292,
430
+ "learning_rate": 3.088888888888889e-05,
431
+ "loss": 0.0834,
432
+ "step": 1450
433
+ },
434
+ {
435
+ "epoch": 1475.0,
436
+ "grad_norm": 3.905423164367676,
437
+ "learning_rate": 2.95e-05,
438
+ "loss": 0.0704,
439
+ "step": 1475
440
+ },
441
+ {
442
+ "epoch": 1500.0,
443
+ "grad_norm": 0.5851226449012756,
444
+ "learning_rate": 2.811111111111111e-05,
445
+ "loss": 0.0938,
446
+ "step": 1500
447
+ },
448
+ {
449
+ "epoch": 1500.0,
450
+ "eval_loss": 0.8552892804145813,
451
+ "eval_runtime": 0.0192,
452
+ "eval_samples_per_second": 52.216,
453
+ "eval_steps_per_second": 52.216,
454
+ "step": 1500
455
+ },
456
+ {
457
+ "epoch": 1525.0,
458
+ "grad_norm": 0.6819572448730469,
459
+ "learning_rate": 2.6722222222222228e-05,
460
+ "loss": 0.0679,
461
+ "step": 1525
462
+ },
463
+ {
464
+ "epoch": 1550.0,
465
+ "grad_norm": 0.800021767616272,
466
+ "learning_rate": 2.5333333333333337e-05,
467
+ "loss": 0.0695,
468
+ "step": 1550
469
+ },
470
+ {
471
+ "epoch": 1575.0,
472
+ "grad_norm": 0.6414686441421509,
473
+ "learning_rate": 2.3944444444444443e-05,
474
+ "loss": 0.069,
475
+ "step": 1575
476
+ },
477
+ {
478
+ "epoch": 1600.0,
479
+ "grad_norm": 0.7216551303863525,
480
+ "learning_rate": 2.255555555555556e-05,
481
+ "loss": 0.0688,
482
+ "step": 1600
483
+ },
484
+ {
485
+ "epoch": 1625.0,
486
+ "grad_norm": 2.296966075897217,
487
+ "learning_rate": 2.116666666666667e-05,
488
+ "loss": 0.0705,
489
+ "step": 1625
490
+ },
491
+ {
492
+ "epoch": 1650.0,
493
+ "grad_norm": 0.7505176067352295,
494
+ "learning_rate": 1.9777777777777778e-05,
495
+ "loss": 0.0788,
496
+ "step": 1650
497
+ },
498
+ {
499
+ "epoch": 1675.0,
500
+ "grad_norm": 0.9566255807876587,
501
+ "learning_rate": 1.838888888888889e-05,
502
+ "loss": 0.0799,
503
+ "step": 1675
504
+ },
505
+ {
506
+ "epoch": 1700.0,
507
+ "grad_norm": 3.685331106185913,
508
+ "learning_rate": 1.7000000000000003e-05,
509
+ "loss": 0.0805,
510
+ "step": 1700
511
+ },
512
+ {
513
+ "epoch": 1725.0,
514
+ "grad_norm": 0.3868086338043213,
515
+ "learning_rate": 1.5611111111111113e-05,
516
+ "loss": 0.075,
517
+ "step": 1725
518
+ },
519
+ {
520
+ "epoch": 1750.0,
521
+ "grad_norm": 0.690929114818573,
522
+ "learning_rate": 1.4222222222222224e-05,
523
+ "loss": 0.0637,
524
+ "step": 1750
525
+ },
526
+ {
527
+ "epoch": 1775.0,
528
+ "grad_norm": 6.572366714477539,
529
+ "learning_rate": 1.2833333333333333e-05,
530
+ "loss": 0.0709,
531
+ "step": 1775
532
+ },
533
+ {
534
+ "epoch": 1800.0,
535
+ "grad_norm": 0.6316695213317871,
536
+ "learning_rate": 1.1444444444444446e-05,
537
+ "loss": 0.0708,
538
+ "step": 1800
539
+ },
540
+ {
541
+ "epoch": 1825.0,
542
+ "grad_norm": 16.782623291015625,
543
+ "learning_rate": 1.0055555555555555e-05,
544
+ "loss": 0.0711,
545
+ "step": 1825
546
+ },
547
+ {
548
+ "epoch": 1850.0,
549
+ "grad_norm": 0.42628633975982666,
550
+ "learning_rate": 8.666666666666668e-06,
551
+ "loss": 0.0713,
552
+ "step": 1850
553
+ },
554
+ {
555
+ "epoch": 1875.0,
556
+ "grad_norm": 0.48583984375,
557
+ "learning_rate": 7.277777777777778e-06,
558
+ "loss": 0.0665,
559
+ "step": 1875
560
+ },
561
+ {
562
+ "epoch": 1900.0,
563
+ "grad_norm": 0.5549265146255493,
564
+ "learning_rate": 5.888888888888889e-06,
565
+ "loss": 0.0658,
566
+ "step": 1900
567
+ },
568
+ {
569
+ "epoch": 1925.0,
570
+ "grad_norm": 2.4642038345336914,
571
+ "learning_rate": 4.5e-06,
572
+ "loss": 0.0687,
573
+ "step": 1925
574
+ },
575
+ {
576
+ "epoch": 1950.0,
577
+ "grad_norm": 0.607486367225647,
578
+ "learning_rate": 3.111111111111111e-06,
579
+ "loss": 0.0733,
580
+ "step": 1950
581
+ },
582
+ {
583
+ "epoch": 1975.0,
584
+ "grad_norm": 0.6398457884788513,
585
+ "learning_rate": 1.7222222222222222e-06,
586
+ "loss": 0.066,
587
+ "step": 1975
588
+ },
589
+ {
590
+ "epoch": 2000.0,
591
+ "grad_norm": 0.43928271532058716,
592
+ "learning_rate": 3.3333333333333335e-07,
593
+ "loss": 0.1062,
594
+ "step": 2000
595
+ },
596
+ {
597
+ "epoch": 2000.0,
598
+ "eval_loss": 0.8336465954780579,
599
+ "eval_runtime": 0.0207,
600
+ "eval_samples_per_second": 48.371,
601
+ "eval_steps_per_second": 48.371,
602
+ "step": 2000
603
+ }
604
+ ],
605
+ "logging_steps": 25,
606
+ "max_steps": 2000,
607
+ "num_input_tokens_seen": 0,
608
+ "num_train_epochs": 2000,
609
+ "save_steps": 500,
610
+ "stateful_callbacks": {
611
+ "TrainerControl": {
612
+ "args": {
613
+ "should_epoch_stop": false,
614
+ "should_evaluate": false,
615
+ "should_log": false,
616
+ "should_save": true,
617
+ "should_training_stop": true
618
+ },
619
+ "attributes": {}
620
+ }
621
+ },
622
+ "total_flos": 48501262656000.0,
623
+ "train_batch_size": 8,
624
+ "trial_name": null,
625
+ "trial_params": null
626
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:86c597e718cbfffd8627917f8791496e241d780ef852c6b9c9e24956387e698d
3
+ size 5905