Hugging Face's logo

gguichard
/
wsd_myriade_synth_data_gpt4turbov4

Model card Files Community
gguichard commited on
Commit
9b1120f
•
1 Parent(s): b92df0b

Training in progress, epoch 32, checkpoint

Browse files
Files changed (12) hide show
  1. checkpoint-6784/README.md +9 -0
  2. checkpoint-6784/adapter_config.json +26 -0
  3. checkpoint-6784/adapter_model.bin +3 -0
  4. checkpoint-6784/optimizer.pt +3 -0
  5. checkpoint-6784/rng_state.pth +3 -0
  6. checkpoint-6784/scheduler.pt +3 -0
  7. checkpoint-6784/sentencepiece.bpe.model +3 -0
  8. checkpoint-6784/special_tokens_map.json +19 -0
  9. checkpoint-6784/tokenizer.json +0 -0
  10. checkpoint-6784/tokenizer_config.json +24 -0
  11. checkpoint-6784/trainer_state.json +467 -0
  12. checkpoint-6784/training_args.bin +3 -0
checkpoint-6784/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-6784/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": 64,
18
+ "revision": null,
19
+ "target_modules": [
20
+ "query",
21
+ "key",
22
+ "value",
23
+ "dense"
24
+ ],
25
+ "task_type": "TOKEN_CLS"
26
+ }
checkpoint-6784/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:abbacc810280334fa8d81c4b053f4daff4125065677bec844045e6d7bf83f0b9
3
+ size 117304221
checkpoint-6784/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:abf2c8f4ac2a5970330cede4891b73afe58aa88cb26a42b132e4b86d242b15a3
3
+ size 234652550
checkpoint-6784/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8df31dcd8a92704a34c89fab11d6011a049892f7d17744a01d99c58673f3b639
3
+ size 14575
checkpoint-6784/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a34b9d71447ba9c3cb9ab3f2b2142714adbf74d2578415d91aaea784bf656b79
3
+ size 627
checkpoint-6784/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-6784/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-6784/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-6784/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-6784/trainer_state.json ADDED
@@ -0,0 +1,467 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 32.0,
5
+ "eval_steps": 500,
6
+ "global_step": 6784,
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": 6.3714,
15
+ "step": 212
16
+ },
17
+ {
18
+ "epoch": 1.0,
19
+ "eval_loss": 5.850927352905273,
20
+ "eval_runtime": 3.8282,
21
+ "eval_samples_per_second": 176.584,
22
+ "eval_steps_per_second": 22.204,
23
+ "step": 212
24
+ },
25
+ {
26
+ "epoch": 2.0,
27
+ "learning_rate": 9.900000000000001e-05,
28
+ "loss": 5.5438,
29
+ "step": 424
30
+ },
31
+ {
32
+ "epoch": 2.0,
33
+ "eval_loss": 4.862037658691406,
34
+ "eval_runtime": 3.7425,
35
+ "eval_samples_per_second": 180.626,
36
+ "eval_steps_per_second": 22.712,
37
+ "step": 424
38
+ },
39
+ {
40
+ "epoch": 3.0,
41
+ "learning_rate": 9.850000000000001e-05,
42
+ "loss": 4.6993,
43
+ "step": 636
44
+ },
45
+ {
46
+ "epoch": 3.0,
47
+ "eval_loss": 4.018372058868408,
48
+ "eval_runtime": 4.0424,
49
+ "eval_samples_per_second": 167.229,
50
+ "eval_steps_per_second": 21.027,
51
+ "step": 636
52
+ },
53
+ {
54
+ "epoch": 4.0,
55
+ "learning_rate": 9.8e-05,
56
+ "loss": 3.967,
57
+ "step": 848
58
+ },
59
+ {
60
+ "epoch": 4.0,
61
+ "eval_loss": 3.290557384490967,
62
+ "eval_runtime": 4.3933,
63
+ "eval_samples_per_second": 153.872,
64
+ "eval_steps_per_second": 19.348,
65
+ "step": 848
66
+ },
67
+ {
68
+ "epoch": 5.0,
69
+ "learning_rate": 9.75e-05,
70
+ "loss": 3.305,
71
+ "step": 1060
72
+ },
73
+ {
74
+ "epoch": 5.0,
75
+ "eval_loss": 2.6497254371643066,
76
+ "eval_runtime": 4.3562,
77
+ "eval_samples_per_second": 155.182,
78
+ "eval_steps_per_second": 19.513,
79
+ "step": 1060
80
+ },
81
+ {
82
+ "epoch": 6.0,
83
+ "learning_rate": 9.7e-05,
84
+ "loss": 2.7185,
85
+ "step": 1272
86
+ },
87
+ {
88
+ "epoch": 6.0,
89
+ "eval_loss": 2.1158342361450195,
90
+ "eval_runtime": 4.5634,
91
+ "eval_samples_per_second": 148.135,
92
+ "eval_steps_per_second": 18.626,
93
+ "step": 1272
94
+ },
95
+ {
96
+ "epoch": 7.0,
97
+ "learning_rate": 9.65e-05,
98
+ "loss": 2.2181,
99
+ "step": 1484
100
+ },
101
+ {
102
+ "epoch": 7.0,
103
+ "eval_loss": 1.6882902383804321,
104
+ "eval_runtime": 4.3232,
105
+ "eval_samples_per_second": 156.367,
106
+ "eval_steps_per_second": 19.662,
107
+ "step": 1484
108
+ },
109
+ {
110
+ "epoch": 8.0,
111
+ "learning_rate": 9.6e-05,
112
+ "loss": 1.8106,
113
+ "step": 1696
114
+ },
115
+ {
116
+ "epoch": 8.0,
117
+ "eval_loss": 1.3730388879776,
118
+ "eval_runtime": 4.3486,
119
+ "eval_samples_per_second": 155.451,
120
+ "eval_steps_per_second": 19.546,
121
+ "step": 1696
122
+ },
123
+ {
124
+ "epoch": 9.0,
125
+ "learning_rate": 9.55e-05,
126
+ "loss": 1.497,
127
+ "step": 1908
128
+ },
129
+ {
130
+ "epoch": 9.0,
131
+ "eval_loss": 1.142008900642395,
132
+ "eval_runtime": 4.2743,
133
+ "eval_samples_per_second": 158.154,
134
+ "eval_steps_per_second": 19.886,
135
+ "step": 1908
136
+ },
137
+ {
138
+ "epoch": 10.0,
139
+ "learning_rate": 9.5e-05,
140
+ "loss": 1.2526,
141
+ "step": 2120
142
+ },
143
+ {
144
+ "epoch": 10.0,
145
+ "eval_loss": 0.9753935933113098,
146
+ "eval_runtime": 4.2134,
147
+ "eval_samples_per_second": 160.44,
148
+ "eval_steps_per_second": 20.174,
149
+ "step": 2120
150
+ },
151
+ {
152
+ "epoch": 11.0,
153
+ "learning_rate": 9.449999999999999e-05,
154
+ "loss": 1.0576,
155
+ "step": 2332
156
+ },
157
+ {
158
+ "epoch": 11.0,
159
+ "eval_loss": 0.8463678956031799,
160
+ "eval_runtime": 4.432,
161
+ "eval_samples_per_second": 152.527,
162
+ "eval_steps_per_second": 19.179,
163
+ "step": 2332
164
+ },
165
+ {
166
+ "epoch": 12.0,
167
+ "learning_rate": 9.4e-05,
168
+ "loss": 0.8965,
169
+ "step": 2544
170
+ },
171
+ {
172
+ "epoch": 12.0,
173
+ "eval_loss": 0.7349700927734375,
174
+ "eval_runtime": 4.5659,
175
+ "eval_samples_per_second": 148.054,
176
+ "eval_steps_per_second": 18.616,
177
+ "step": 2544
178
+ },
179
+ {
180
+ "epoch": 13.0,
181
+ "learning_rate": 9.350000000000001e-05,
182
+ "loss": 0.7613,
183
+ "step": 2756
184
+ },
185
+ {
186
+ "epoch": 13.0,
187
+ "eval_loss": 0.6494072675704956,
188
+ "eval_runtime": 4.499,
189
+ "eval_samples_per_second": 150.254,
190
+ "eval_steps_per_second": 18.893,
191
+ "step": 2756
192
+ },
193
+ {
194
+ "epoch": 14.0,
195
+ "learning_rate": 9.300000000000001e-05,
196
+ "loss": 0.645,
197
+ "step": 2968
198
+ },
199
+ {
200
+ "epoch": 14.0,
201
+ "eval_loss": 0.5844011902809143,
202
+ "eval_runtime": 4.7455,
203
+ "eval_samples_per_second": 142.449,
204
+ "eval_steps_per_second": 17.912,
205
+ "step": 2968
206
+ },
207
+ {
208
+ "epoch": 15.0,
209
+ "learning_rate": 9.250000000000001e-05,
210
+ "loss": 0.548,
211
+ "step": 3180
212
+ },
213
+ {
214
+ "epoch": 15.0,
215
+ "eval_loss": 0.5139679312705994,
216
+ "eval_runtime": 4.7986,
217
+ "eval_samples_per_second": 140.876,
218
+ "eval_steps_per_second": 17.714,
219
+ "step": 3180
220
+ },
221
+ {
222
+ "epoch": 16.0,
223
+ "learning_rate": 9.200000000000001e-05,
224
+ "loss": 0.4564,
225
+ "step": 3392
226
+ },
227
+ {
228
+ "epoch": 16.0,
229
+ "eval_loss": 0.47241923213005066,
230
+ "eval_runtime": 4.6502,
231
+ "eval_samples_per_second": 145.371,
232
+ "eval_steps_per_second": 18.279,
233
+ "step": 3392
234
+ },
235
+ {
236
+ "epoch": 17.0,
237
+ "learning_rate": 9.15e-05,
238
+ "loss": 0.3785,
239
+ "step": 3604
240
+ },
241
+ {
242
+ "epoch": 17.0,
243
+ "eval_loss": 0.4442097246646881,
244
+ "eval_runtime": 4.6864,
245
+ "eval_samples_per_second": 144.247,
246
+ "eval_steps_per_second": 18.138,
247
+ "step": 3604
248
+ },
249
+ {
250
+ "epoch": 18.0,
251
+ "learning_rate": 9.1e-05,
252
+ "loss": 0.3144,
253
+ "step": 3816
254
+ },
255
+ {
256
+ "epoch": 18.0,
257
+ "eval_loss": 0.407890647649765,
258
+ "eval_runtime": 4.6905,
259
+ "eval_samples_per_second": 144.121,
260
+ "eval_steps_per_second": 18.122,
261
+ "step": 3816
262
+ },
263
+ {
264
+ "epoch": 19.0,
265
+ "learning_rate": 9.05e-05,
266
+ "loss": 0.2609,
267
+ "step": 4028
268
+ },
269
+ {
270
+ "epoch": 19.0,
271
+ "eval_loss": 0.3970014452934265,
272
+ "eval_runtime": 4.6686,
273
+ "eval_samples_per_second": 144.799,
274
+ "eval_steps_per_second": 18.207,
275
+ "step": 4028
276
+ },
277
+ {
278
+ "epoch": 20.0,
279
+ "learning_rate": 9e-05,
280
+ "loss": 0.2106,
281
+ "step": 4240
282
+ },
283
+ {
284
+ "epoch": 20.0,
285
+ "eval_loss": 0.37661823630332947,
286
+ "eval_runtime": 4.7062,
287
+ "eval_samples_per_second": 143.641,
288
+ "eval_steps_per_second": 18.061,
289
+ "step": 4240
290
+ },
291
+ {
292
+ "epoch": 21.0,
293
+ "learning_rate": 8.950000000000001e-05,
294
+ "loss": 0.1727,
295
+ "step": 4452
296
+ },
297
+ {
298
+ "epoch": 21.0,
299
+ "eval_loss": 0.37585482001304626,
300
+ "eval_runtime": 4.8302,
301
+ "eval_samples_per_second": 139.953,
302
+ "eval_steps_per_second": 17.598,
303
+ "step": 4452
304
+ },
305
+ {
306
+ "epoch": 22.0,
307
+ "learning_rate": 8.900000000000001e-05,
308
+ "loss": 0.1427,
309
+ "step": 4664
310
+ },
311
+ {
312
+ "epoch": 22.0,
313
+ "eval_loss": 0.35890331864356995,
314
+ "eval_runtime": 4.823,
315
+ "eval_samples_per_second": 140.16,
316
+ "eval_steps_per_second": 17.624,
317
+ "step": 4664
318
+ },
319
+ {
320
+ "epoch": 23.0,
321
+ "learning_rate": 8.850000000000001e-05,
322
+ "loss": 0.1115,
323
+ "step": 4876
324
+ },
325
+ {
326
+ "epoch": 23.0,
327
+ "eval_loss": 0.35352450609207153,
328
+ "eval_runtime": 4.9556,
329
+ "eval_samples_per_second": 136.411,
330
+ "eval_steps_per_second": 17.152,
331
+ "step": 4876
332
+ },
333
+ {
334
+ "epoch": 24.0,
335
+ "learning_rate": 8.800000000000001e-05,
336
+ "loss": 0.0949,
337
+ "step": 5088
338
+ },
339
+ {
340
+ "epoch": 24.0,
341
+ "eval_loss": 0.3508472442626953,
342
+ "eval_runtime": 4.4886,
343
+ "eval_samples_per_second": 150.605,
344
+ "eval_steps_per_second": 18.937,
345
+ "step": 5088
346
+ },
347
+ {
348
+ "epoch": 25.0,
349
+ "learning_rate": 8.75e-05,
350
+ "loss": 0.0779,
351
+ "step": 5300
352
+ },
353
+ {
354
+ "epoch": 25.0,
355
+ "eval_loss": 0.35696595907211304,
356
+ "eval_runtime": 4.5496,
357
+ "eval_samples_per_second": 148.586,
358
+ "eval_steps_per_second": 18.683,
359
+ "step": 5300
360
+ },
361
+ {
362
+ "epoch": 26.0,
363
+ "learning_rate": 8.7e-05,
364
+ "loss": 0.0652,
365
+ "step": 5512
366
+ },
367
+ {
368
+ "epoch": 26.0,
369
+ "eval_loss": 0.3560032546520233,
370
+ "eval_runtime": 5.7794,
371
+ "eval_samples_per_second": 116.966,
372
+ "eval_steps_per_second": 14.707,
373
+ "step": 5512
374
+ },
375
+ {
376
+ "epoch": 27.0,
377
+ "learning_rate": 8.65e-05,
378
+ "loss": 0.0549,
379
+ "step": 5724
380
+ },
381
+ {
382
+ "epoch": 27.0,
383
+ "eval_loss": 0.355691134929657,
384
+ "eval_runtime": 5.1711,
385
+ "eval_samples_per_second": 130.726,
386
+ "eval_steps_per_second": 16.437,
387
+ "step": 5724
388
+ },
389
+ {
390
+ "epoch": 28.0,
391
+ "learning_rate": 8.6e-05,
392
+ "loss": 0.0431,
393
+ "step": 5936
394
+ },
395
+ {
396
+ "epoch": 28.0,
397
+ "eval_loss": 0.3634819984436035,
398
+ "eval_runtime": 4.8044,
399
+ "eval_samples_per_second": 140.705,
400
+ "eval_steps_per_second": 17.692,
401
+ "step": 5936
402
+ },
403
+ {
404
+ "epoch": 29.0,
405
+ "learning_rate": 8.55e-05,
406
+ "loss": 0.0389,
407
+ "step": 6148
408
+ },
409
+ {
410
+ "epoch": 29.0,
411
+ "eval_loss": 0.36610594391822815,
412
+ "eval_runtime": 4.5861,
413
+ "eval_samples_per_second": 147.403,
414
+ "eval_steps_per_second": 18.534,
415
+ "step": 6148
416
+ },
417
+ {
418
+ "epoch": 30.0,
419
+ "learning_rate": 8.5e-05,
420
+ "loss": 0.0327,
421
+ "step": 6360
422
+ },
423
+ {
424
+ "epoch": 30.0,
425
+ "eval_loss": 0.3725152611732483,
426
+ "eval_runtime": 5.6392,
427
+ "eval_samples_per_second": 119.874,
428
+ "eval_steps_per_second": 15.073,
429
+ "step": 6360
430
+ },
431
+ {
432
+ "epoch": 31.0,
433
+ "learning_rate": 8.450000000000001e-05,
434
+ "loss": 0.0293,
435
+ "step": 6572
436
+ },
437
+ {
438
+ "epoch": 31.0,
439
+ "eval_loss": 0.38449323177337646,
440
+ "eval_runtime": 5.8725,
441
+ "eval_samples_per_second": 115.113,
442
+ "eval_steps_per_second": 14.474,
443
+ "step": 6572
444
+ },
445
+ {
446
+ "epoch": 32.0,
447
+ "learning_rate": 8.4e-05,
448
+ "loss": 0.0279,
449
+ "step": 6784
450
+ },
451
+ {
452
+ "epoch": 32.0,
453
+ "eval_loss": 0.3865319490432739,
454
+ "eval_runtime": 4.6302,
455
+ "eval_samples_per_second": 145.997,
456
+ "eval_steps_per_second": 18.358,
457
+ "step": 6784
458
+ }
459
+ ],
460
+ "logging_steps": 500,
461
+ "max_steps": 42400,
462
+ "num_train_epochs": 200,
463
+ "save_steps": 500,
464
+ "total_flos": 9132817466469660.0,
465
+ "trial_name": null,
466
+ "trial_params": null
467
+ }
checkpoint-6784/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e7d47d9ef4da320ee46463f8716d365eb34b4bc5488090164af1bcf3ad60b35
3
+ size 4155