Hugging Face's logo

gguichard
/
wsd_myriade_synth_data_gpt4turbov3

Model card Files Community
gguichard commited on
Commit
af03f75
•
1 Parent(s): 5024883

Training in progress, epoch 29, checkpoint

Browse files
Files changed (12) hide show
  1. checkpoint-6148/README.md +9 -0
  2. checkpoint-6148/adapter_config.json +26 -0
  3. checkpoint-6148/adapter_model.bin +3 -0
  4. checkpoint-6148/optimizer.pt +3 -0
  5. checkpoint-6148/rng_state.pth +3 -0
  6. checkpoint-6148/scheduler.pt +3 -0
  7. checkpoint-6148/sentencepiece.bpe.model +3 -0
  8. checkpoint-6148/special_tokens_map.json +19 -0
  9. checkpoint-6148/tokenizer.json +0 -0
  10. checkpoint-6148/tokenizer_config.json +24 -0
  11. checkpoint-6148/trainer_state.json +425 -0
  12. checkpoint-6148/training_args.bin +3 -0
checkpoint-6148/README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+
9
+ - PEFT 0.5.0
checkpoint-6148/adapter_config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "camembert/camembert-large",
4
+ "bias": "all",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 16,
11
+ "lora_dropout": 0.1,
12
+ "modules_to_save": [
13
+ "classifier.bias",
14
+ "classifier.weight"
15
+ ],
16
+ "peft_type": "LORA",
17
+ "r": 16,
18
+ "revision": null,
19
+ "target_modules": [
20
+ "query",
21
+ "key",
22
+ "value",
23
+ "dense"
24
+ ],
25
+ "task_type": "TOKEN_CLS"
26
+ }
checkpoint-6148/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6c7b14ac40411640523ab53ad31906ea4968fe34e89a5d853fb38f2663f45549
3
+ size 32369117
checkpoint-6148/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4c35fda5589df5169ac44cc373c2a50a2fcae7b43962a155c914eb8cfae4cdb5
3
+ size 64782278
checkpoint-6148/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:93b1fd4e78c7176cc40d1a4fc29743439b340df6c7fccc61a18a1d6b83aff449
3
+ size 14575
checkpoint-6148/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3761cbc6de9504b7e28226506f8f35ee48dac4d50312b91eae75879db88f5db6
3
+ size 627
checkpoint-6148/sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f98f266fdc548c94216aaadc13ffaaafacf0c8793303e2195322d954549ea261
3
+ size 808767
checkpoint-6148/special_tokens_map.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<s>NOTUSED",
4
+ "</s>NOTUSED"
5
+ ],
6
+ "bos_token": "<s>",
7
+ "cls_token": "<s>",
8
+ "eos_token": "</s>",
9
+ "mask_token": {
10
+ "content": "<mask>",
11
+ "lstrip": true,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "<pad>",
17
+ "sep_token": "</s>",
18
+ "unk_token": "<unk>"
19
+ }
checkpoint-6148/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-6148/tokenizer_config.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<s>NOTUSED",
4
+ "</s>NOTUSED"
5
+ ],
6
+ "bos_token": "<s>",
7
+ "clean_up_tokenization_spaces": true,
8
+ "cls_token": "<s>",
9
+ "eos_token": "</s>",
10
+ "mask_token": {
11
+ "__type": "AddedToken",
12
+ "content": "<mask>",
13
+ "lstrip": true,
14
+ "normalized": true,
15
+ "rstrip": false,
16
+ "single_word": false
17
+ },
18
+ "model_max_length": 1000000000000000019884624838656,
19
+ "pad_token": "<pad>",
20
+ "sep_token": "</s>",
21
+ "sp_model_kwargs": {},
22
+ "tokenizer_class": "CamembertTokenizer",
23
+ "unk_token": "<unk>"
24
+ }
checkpoint-6148/trainer_state.json ADDED
@@ -0,0 +1,425 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 29.0,
5
+ "eval_steps": 500,
6
+ "global_step": 6148,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 1.0,
13
+ "learning_rate": 9.95e-05,
14
+ "loss": 6.3597,
15
+ "step": 212
16
+ },
17
+ {
18
+ "epoch": 1.0,
19
+ "eval_loss": 5.853069305419922,
20
+ "eval_runtime": 3.5937,
21
+ "eval_samples_per_second": 188.106,
22
+ "eval_steps_per_second": 23.652,
23
+ "step": 212
24
+ },
25
+ {
26
+ "epoch": 2.0,
27
+ "learning_rate": 9.900000000000001e-05,
28
+ "loss": 5.5629,
29
+ "step": 424
30
+ },
31
+ {
32
+ "epoch": 2.0,
33
+ "eval_loss": 4.915834903717041,
34
+ "eval_runtime": 3.704,
35
+ "eval_samples_per_second": 182.506,
36
+ "eval_steps_per_second": 22.948,
37
+ "step": 424
38
+ },
39
+ {
40
+ "epoch": 3.0,
41
+ "learning_rate": 9.850000000000001e-05,
42
+ "loss": 4.8228,
43
+ "step": 636
44
+ },
45
+ {
46
+ "epoch": 3.0,
47
+ "eval_loss": 4.177055835723877,
48
+ "eval_runtime": 3.7104,
49
+ "eval_samples_per_second": 182.193,
50
+ "eval_steps_per_second": 22.909,
51
+ "step": 636
52
+ },
53
+ {
54
+ "epoch": 4.0,
55
+ "learning_rate": 9.8e-05,
56
+ "loss": 4.1035,
57
+ "step": 848
58
+ },
59
+ {
60
+ "epoch": 4.0,
61
+ "eval_loss": 3.462527275085449,
62
+ "eval_runtime": 3.1205,
63
+ "eval_samples_per_second": 216.633,
64
+ "eval_steps_per_second": 27.239,
65
+ "step": 848
66
+ },
67
+ {
68
+ "epoch": 5.0,
69
+ "learning_rate": 9.75e-05,
70
+ "loss": 3.467,
71
+ "step": 1060
72
+ },
73
+ {
74
+ "epoch": 5.0,
75
+ "eval_loss": 2.8398332595825195,
76
+ "eval_runtime": 3.3235,
77
+ "eval_samples_per_second": 203.402,
78
+ "eval_steps_per_second": 25.576,
79
+ "step": 1060
80
+ },
81
+ {
82
+ "epoch": 6.0,
83
+ "learning_rate": 9.7e-05,
84
+ "loss": 2.8887,
85
+ "step": 1272
86
+ },
87
+ {
88
+ "epoch": 6.0,
89
+ "eval_loss": 2.3201420307159424,
90
+ "eval_runtime": 3.4255,
91
+ "eval_samples_per_second": 197.342,
92
+ "eval_steps_per_second": 24.814,
93
+ "step": 1272
94
+ },
95
+ {
96
+ "epoch": 7.0,
97
+ "learning_rate": 9.65e-05,
98
+ "loss": 2.4003,
99
+ "step": 1484
100
+ },
101
+ {
102
+ "epoch": 7.0,
103
+ "eval_loss": 1.8698177337646484,
104
+ "eval_runtime": 3.5218,
105
+ "eval_samples_per_second": 191.945,
106
+ "eval_steps_per_second": 24.135,
107
+ "step": 1484
108
+ },
109
+ {
110
+ "epoch": 8.0,
111
+ "learning_rate": 9.6e-05,
112
+ "loss": 1.9919,
113
+ "step": 1696
114
+ },
115
+ {
116
+ "epoch": 8.0,
117
+ "eval_loss": 1.532776951789856,
118
+ "eval_runtime": 3.5514,
119
+ "eval_samples_per_second": 190.345,
120
+ "eval_steps_per_second": 23.934,
121
+ "step": 1696
122
+ },
123
+ {
124
+ "epoch": 9.0,
125
+ "learning_rate": 9.55e-05,
126
+ "loss": 1.6656,
127
+ "step": 1908
128
+ },
129
+ {
130
+ "epoch": 9.0,
131
+ "eval_loss": 1.2841078042984009,
132
+ "eval_runtime": 3.622,
133
+ "eval_samples_per_second": 186.637,
134
+ "eval_steps_per_second": 23.468,
135
+ "step": 1908
136
+ },
137
+ {
138
+ "epoch": 10.0,
139
+ "learning_rate": 9.5e-05,
140
+ "loss": 1.4036,
141
+ "step": 2120
142
+ },
143
+ {
144
+ "epoch": 10.0,
145
+ "eval_loss": 1.107720971107483,
146
+ "eval_runtime": 3.6471,
147
+ "eval_samples_per_second": 185.353,
148
+ "eval_steps_per_second": 23.306,
149
+ "step": 2120
150
+ },
151
+ {
152
+ "epoch": 11.0,
153
+ "learning_rate": 9.449999999999999e-05,
154
+ "loss": 1.2091,
155
+ "step": 2332
156
+ },
157
+ {
158
+ "epoch": 11.0,
159
+ "eval_loss": 0.9709760546684265,
160
+ "eval_runtime": 3.7402,
161
+ "eval_samples_per_second": 180.741,
162
+ "eval_steps_per_second": 22.726,
163
+ "step": 2332
164
+ },
165
+ {
166
+ "epoch": 12.0,
167
+ "learning_rate": 9.4e-05,
168
+ "loss": 1.0379,
169
+ "step": 2544
170
+ },
171
+ {
172
+ "epoch": 12.0,
173
+ "eval_loss": 0.8528622984886169,
174
+ "eval_runtime": 3.734,
175
+ "eval_samples_per_second": 181.041,
176
+ "eval_steps_per_second": 22.764,
177
+ "step": 2544
178
+ },
179
+ {
180
+ "epoch": 13.0,
181
+ "learning_rate": 9.350000000000001e-05,
182
+ "loss": 0.8994,
183
+ "step": 2756
184
+ },
185
+ {
186
+ "epoch": 13.0,
187
+ "eval_loss": 0.7617917060852051,
188
+ "eval_runtime": 3.7153,
189
+ "eval_samples_per_second": 181.95,
190
+ "eval_steps_per_second": 22.878,
191
+ "step": 2756
192
+ },
193
+ {
194
+ "epoch": 14.0,
195
+ "learning_rate": 9.300000000000001e-05,
196
+ "loss": 0.7828,
197
+ "step": 2968
198
+ },
199
+ {
200
+ "epoch": 14.0,
201
+ "eval_loss": 0.6774648427963257,
202
+ "eval_runtime": 3.6994,
203
+ "eval_samples_per_second": 182.734,
204
+ "eval_steps_per_second": 22.977,
205
+ "step": 2968
206
+ },
207
+ {
208
+ "epoch": 15.0,
209
+ "learning_rate": 9.250000000000001e-05,
210
+ "loss": 0.6788,
211
+ "step": 3180
212
+ },
213
+ {
214
+ "epoch": 15.0,
215
+ "eval_loss": 0.6361714005470276,
216
+ "eval_runtime": 3.6581,
217
+ "eval_samples_per_second": 184.796,
218
+ "eval_steps_per_second": 23.236,
219
+ "step": 3180
220
+ },
221
+ {
222
+ "epoch": 16.0,
223
+ "learning_rate": 9.200000000000001e-05,
224
+ "loss": 0.5819,
225
+ "step": 3392
226
+ },
227
+ {
228
+ "epoch": 16.0,
229
+ "eval_loss": 0.5761357545852661,
230
+ "eval_runtime": 3.7378,
231
+ "eval_samples_per_second": 180.853,
232
+ "eval_steps_per_second": 22.74,
233
+ "step": 3392
234
+ },
235
+ {
236
+ "epoch": 17.0,
237
+ "learning_rate": 9.15e-05,
238
+ "loss": 0.5104,
239
+ "step": 3604
240
+ },
241
+ {
242
+ "epoch": 17.0,
243
+ "eval_loss": 0.5219044089317322,
244
+ "eval_runtime": 3.7453,
245
+ "eval_samples_per_second": 180.495,
246
+ "eval_steps_per_second": 22.695,
247
+ "step": 3604
248
+ },
249
+ {
250
+ "epoch": 18.0,
251
+ "learning_rate": 9.1e-05,
252
+ "loss": 0.4343,
253
+ "step": 3816
254
+ },
255
+ {
256
+ "epoch": 18.0,
257
+ "eval_loss": 0.4844624102115631,
258
+ "eval_runtime": 3.7695,
259
+ "eval_samples_per_second": 179.335,
260
+ "eval_steps_per_second": 22.55,
261
+ "step": 3816
262
+ },
263
+ {
264
+ "epoch": 19.0,
265
+ "learning_rate": 9.05e-05,
266
+ "loss": 0.3784,
267
+ "step": 4028
268
+ },
269
+ {
270
+ "epoch": 19.0,
271
+ "eval_loss": 0.44724151492118835,
272
+ "eval_runtime": 3.7703,
273
+ "eval_samples_per_second": 179.296,
274
+ "eval_steps_per_second": 22.545,
275
+ "step": 4028
276
+ },
277
+ {
278
+ "epoch": 20.0,
279
+ "learning_rate": 9e-05,
280
+ "loss": 0.3224,
281
+ "step": 4240
282
+ },
283
+ {
284
+ "epoch": 20.0,
285
+ "eval_loss": 0.45848363637924194,
286
+ "eval_runtime": 3.6822,
287
+ "eval_samples_per_second": 183.584,
288
+ "eval_steps_per_second": 23.084,
289
+ "step": 4240
290
+ },
291
+ {
292
+ "epoch": 21.0,
293
+ "learning_rate": 8.950000000000001e-05,
294
+ "loss": 0.2768,
295
+ "step": 4452
296
+ },
297
+ {
298
+ "epoch": 21.0,
299
+ "eval_loss": 0.4251560866832733,
300
+ "eval_runtime": 3.9348,
301
+ "eval_samples_per_second": 171.802,
302
+ "eval_steps_per_second": 21.602,
303
+ "step": 4452
304
+ },
305
+ {
306
+ "epoch": 22.0,
307
+ "learning_rate": 8.900000000000001e-05,
308
+ "loss": 0.2269,
309
+ "step": 4664
310
+ },
311
+ {
312
+ "epoch": 22.0,
313
+ "eval_loss": 0.39140933752059937,
314
+ "eval_runtime": 3.6907,
315
+ "eval_samples_per_second": 183.165,
316
+ "eval_steps_per_second": 23.031,
317
+ "step": 4664
318
+ },
319
+ {
320
+ "epoch": 23.0,
321
+ "learning_rate": 8.850000000000001e-05,
322
+ "loss": 0.1868,
323
+ "step": 4876
324
+ },
325
+ {
326
+ "epoch": 23.0,
327
+ "eval_loss": 0.4142400920391083,
328
+ "eval_runtime": 3.8669,
329
+ "eval_samples_per_second": 174.815,
330
+ "eval_steps_per_second": 21.981,
331
+ "step": 4876
332
+ },
333
+ {
334
+ "epoch": 24.0,
335
+ "learning_rate": 8.800000000000001e-05,
336
+ "loss": 0.1539,
337
+ "step": 5088
338
+ },
339
+ {
340
+ "epoch": 24.0,
341
+ "eval_loss": 0.3828276991844177,
342
+ "eval_runtime": 3.7873,
343
+ "eval_samples_per_second": 178.493,
344
+ "eval_steps_per_second": 22.444,
345
+ "step": 5088
346
+ },
347
+ {
348
+ "epoch": 25.0,
349
+ "learning_rate": 8.75e-05,
350
+ "loss": 0.1298,
351
+ "step": 5300
352
+ },
353
+ {
354
+ "epoch": 25.0,
355
+ "eval_loss": 0.40518617630004883,
356
+ "eval_runtime": 3.9459,
357
+ "eval_samples_per_second": 171.319,
358
+ "eval_steps_per_second": 21.542,
359
+ "step": 5300
360
+ },
361
+ {
362
+ "epoch": 26.0,
363
+ "learning_rate": 8.7e-05,
364
+ "loss": 0.1098,
365
+ "step": 5512
366
+ },
367
+ {
368
+ "epoch": 26.0,
369
+ "eval_loss": 0.420820027589798,
370
+ "eval_runtime": 3.7042,
371
+ "eval_samples_per_second": 182.497,
372
+ "eval_steps_per_second": 22.947,
373
+ "step": 5512
374
+ },
375
+ {
376
+ "epoch": 27.0,
377
+ "learning_rate": 8.65e-05,
378
+ "loss": 0.0905,
379
+ "step": 5724
380
+ },
381
+ {
382
+ "epoch": 27.0,
383
+ "eval_loss": 0.39528989791870117,
384
+ "eval_runtime": 3.784,
385
+ "eval_samples_per_second": 178.645,
386
+ "eval_steps_per_second": 22.463,
387
+ "step": 5724
388
+ },
389
+ {
390
+ "epoch": 28.0,
391
+ "learning_rate": 8.6e-05,
392
+ "loss": 0.0731,
393
+ "step": 5936
394
+ },
395
+ {
396
+ "epoch": 28.0,
397
+ "eval_loss": 0.41354966163635254,
398
+ "eval_runtime": 3.7338,
399
+ "eval_samples_per_second": 181.047,
400
+ "eval_steps_per_second": 22.765,
401
+ "step": 5936
402
+ },
403
+ {
404
+ "epoch": 29.0,
405
+ "learning_rate": 8.55e-05,
406
+ "loss": 0.063,
407
+ "step": 6148
408
+ },
409
+ {
410
+ "epoch": 29.0,
411
+ "eval_loss": 0.39981818199157715,
412
+ "eval_runtime": 3.6841,
413
+ "eval_samples_per_second": 183.489,
414
+ "eval_steps_per_second": 23.072,
415
+ "step": 6148
416
+ }
417
+ ],
418
+ "logging_steps": 500,
419
+ "max_steps": 42400,
420
+ "num_train_epochs": 200,
421
+ "save_steps": 500,
422
+ "total_flos": 7745313230711328.0,
423
+ "trial_name": null,
424
+ "trial_params": null
425
+ }
checkpoint-6148/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8bbad506d5d7c9f6e6c369455b23a5312c048f6fdad38f9f3772905289326b4f
3
+ size 4155