gguichard commited on
Commit
5f089c6
1 Parent(s): 5c793a7

Training in progress, epoch 51, checkpoint

Browse files
checkpoint-10812/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-10812/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-10812/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:09b4359ffb8e4041fc4ad002e962d1bf0806db81fcd38b620643067f64e2267d
3
+ size 32369117
checkpoint-10812/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfe0e6338957423b0d4930defd6040b880d34879b915f5abb1570b69ceb205b1
3
+ size 64782278
checkpoint-10812/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1e8d6f8bf61bf27e965a6100936905a4e1425cec5fd6f196c2fe01905edb1f92
3
+ size 14575
checkpoint-10812/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0aa2748863e70a37f72fe6eb5696da70acd3e0aba5bbb2d9ab390b47fc7c48a8
3
+ size 627
checkpoint-10812/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-10812/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-10812/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-10812/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-10812/trainer_state.json ADDED
@@ -0,0 +1,733 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 51.0,
5
+ "eval_steps": 500,
6
+ "global_step": 10812,
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
+ "epoch": 30.0,
419
+ "learning_rate": 8.5e-05,
420
+ "loss": 0.0543,
421
+ "step": 6360
422
+ },
423
+ {
424
+ "epoch": 30.0,
425
+ "eval_loss": 0.4148443043231964,
426
+ "eval_runtime": 3.9507,
427
+ "eval_samples_per_second": 171.111,
428
+ "eval_steps_per_second": 21.515,
429
+ "step": 6360
430
+ },
431
+ {
432
+ "epoch": 31.0,
433
+ "learning_rate": 8.450000000000001e-05,
434
+ "loss": 0.0452,
435
+ "step": 6572
436
+ },
437
+ {
438
+ "epoch": 31.0,
439
+ "eval_loss": 0.4315831959247589,
440
+ "eval_runtime": 4.0598,
441
+ "eval_samples_per_second": 166.511,
442
+ "eval_steps_per_second": 20.937,
443
+ "step": 6572
444
+ },
445
+ {
446
+ "epoch": 32.0,
447
+ "learning_rate": 8.4e-05,
448
+ "loss": 0.0411,
449
+ "step": 6784
450
+ },
451
+ {
452
+ "epoch": 32.0,
453
+ "eval_loss": 0.4563623070716858,
454
+ "eval_runtime": 3.7904,
455
+ "eval_samples_per_second": 178.346,
456
+ "eval_steps_per_second": 22.425,
457
+ "step": 6784
458
+ },
459
+ {
460
+ "epoch": 33.0,
461
+ "learning_rate": 8.35e-05,
462
+ "loss": 0.0342,
463
+ "step": 6996
464
+ },
465
+ {
466
+ "epoch": 33.0,
467
+ "eval_loss": 0.4423569142818451,
468
+ "eval_runtime": 3.6613,
469
+ "eval_samples_per_second": 184.635,
470
+ "eval_steps_per_second": 23.216,
471
+ "step": 6996
472
+ },
473
+ {
474
+ "epoch": 34.0,
475
+ "learning_rate": 8.3e-05,
476
+ "loss": 0.0285,
477
+ "step": 7208
478
+ },
479
+ {
480
+ "epoch": 34.0,
481
+ "eval_loss": 0.45883628726005554,
482
+ "eval_runtime": 3.8854,
483
+ "eval_samples_per_second": 173.985,
484
+ "eval_steps_per_second": 21.877,
485
+ "step": 7208
486
+ },
487
+ {
488
+ "epoch": 35.0,
489
+ "learning_rate": 8.25e-05,
490
+ "loss": 0.0261,
491
+ "step": 7420
492
+ },
493
+ {
494
+ "epoch": 35.0,
495
+ "eval_loss": 0.4722530245780945,
496
+ "eval_runtime": 3.9505,
497
+ "eval_samples_per_second": 171.117,
498
+ "eval_steps_per_second": 21.516,
499
+ "step": 7420
500
+ },
501
+ {
502
+ "epoch": 36.0,
503
+ "learning_rate": 8.2e-05,
504
+ "loss": 0.0261,
505
+ "step": 7632
506
+ },
507
+ {
508
+ "epoch": 36.0,
509
+ "eval_loss": 0.4807146191596985,
510
+ "eval_runtime": 3.7918,
511
+ "eval_samples_per_second": 178.279,
512
+ "eval_steps_per_second": 22.417,
513
+ "step": 7632
514
+ },
515
+ {
516
+ "epoch": 37.0,
517
+ "learning_rate": 8.15e-05,
518
+ "loss": 0.0199,
519
+ "step": 7844
520
+ },
521
+ {
522
+ "epoch": 37.0,
523
+ "eval_loss": 0.46007126569747925,
524
+ "eval_runtime": 3.8954,
525
+ "eval_samples_per_second": 173.536,
526
+ "eval_steps_per_second": 21.82,
527
+ "step": 7844
528
+ },
529
+ {
530
+ "epoch": 38.0,
531
+ "learning_rate": 8.1e-05,
532
+ "loss": 0.0169,
533
+ "step": 8056
534
+ },
535
+ {
536
+ "epoch": 38.0,
537
+ "eval_loss": 0.4729515314102173,
538
+ "eval_runtime": 3.8647,
539
+ "eval_samples_per_second": 174.918,
540
+ "eval_steps_per_second": 21.994,
541
+ "step": 8056
542
+ },
543
+ {
544
+ "epoch": 39.0,
545
+ "learning_rate": 8.05e-05,
546
+ "loss": 0.0164,
547
+ "step": 8268
548
+ },
549
+ {
550
+ "epoch": 39.0,
551
+ "eval_loss": 0.4991854131221771,
552
+ "eval_runtime": 3.8642,
553
+ "eval_samples_per_second": 174.937,
554
+ "eval_steps_per_second": 21.997,
555
+ "step": 8268
556
+ },
557
+ {
558
+ "epoch": 40.0,
559
+ "learning_rate": 8e-05,
560
+ "loss": 0.0152,
561
+ "step": 8480
562
+ },
563
+ {
564
+ "epoch": 40.0,
565
+ "eval_loss": 0.48734694719314575,
566
+ "eval_runtime": 3.8857,
567
+ "eval_samples_per_second": 173.971,
568
+ "eval_steps_per_second": 21.875,
569
+ "step": 8480
570
+ },
571
+ {
572
+ "epoch": 41.0,
573
+ "learning_rate": 7.950000000000001e-05,
574
+ "loss": 0.0109,
575
+ "step": 8692
576
+ },
577
+ {
578
+ "epoch": 41.0,
579
+ "eval_loss": 0.49513980746269226,
580
+ "eval_runtime": 3.8249,
581
+ "eval_samples_per_second": 176.736,
582
+ "eval_steps_per_second": 22.223,
583
+ "step": 8692
584
+ },
585
+ {
586
+ "epoch": 42.0,
587
+ "learning_rate": 7.900000000000001e-05,
588
+ "loss": 0.0132,
589
+ "step": 8904
590
+ },
591
+ {
592
+ "epoch": 42.0,
593
+ "eval_loss": 0.4910328984260559,
594
+ "eval_runtime": 3.823,
595
+ "eval_samples_per_second": 176.826,
596
+ "eval_steps_per_second": 22.234,
597
+ "step": 8904
598
+ },
599
+ {
600
+ "epoch": 43.0,
601
+ "learning_rate": 7.850000000000001e-05,
602
+ "loss": 0.0086,
603
+ "step": 9116
604
+ },
605
+ {
606
+ "epoch": 43.0,
607
+ "eval_loss": 0.5066259503364563,
608
+ "eval_runtime": 3.8604,
609
+ "eval_samples_per_second": 175.11,
610
+ "eval_steps_per_second": 22.018,
611
+ "step": 9116
612
+ },
613
+ {
614
+ "epoch": 44.0,
615
+ "learning_rate": 7.800000000000001e-05,
616
+ "loss": 0.0084,
617
+ "step": 9328
618
+ },
619
+ {
620
+ "epoch": 44.0,
621
+ "eval_loss": 0.5218311548233032,
622
+ "eval_runtime": 3.8685,
623
+ "eval_samples_per_second": 174.744,
624
+ "eval_steps_per_second": 21.972,
625
+ "step": 9328
626
+ },
627
+ {
628
+ "epoch": 45.0,
629
+ "learning_rate": 7.75e-05,
630
+ "loss": 0.0082,
631
+ "step": 9540
632
+ },
633
+ {
634
+ "epoch": 45.0,
635
+ "eval_loss": 0.5277370810508728,
636
+ "eval_runtime": 3.8269,
637
+ "eval_samples_per_second": 176.644,
638
+ "eval_steps_per_second": 22.211,
639
+ "step": 9540
640
+ },
641
+ {
642
+ "epoch": 46.0,
643
+ "learning_rate": 7.7e-05,
644
+ "loss": 0.0094,
645
+ "step": 9752
646
+ },
647
+ {
648
+ "epoch": 46.0,
649
+ "eval_loss": 0.5367192029953003,
650
+ "eval_runtime": 3.8012,
651
+ "eval_samples_per_second": 177.838,
652
+ "eval_steps_per_second": 22.361,
653
+ "step": 9752
654
+ },
655
+ {
656
+ "epoch": 47.0,
657
+ "learning_rate": 7.65e-05,
658
+ "loss": 0.0117,
659
+ "step": 9964
660
+ },
661
+ {
662
+ "epoch": 47.0,
663
+ "eval_loss": 0.5425814390182495,
664
+ "eval_runtime": 3.8527,
665
+ "eval_samples_per_second": 175.46,
666
+ "eval_steps_per_second": 22.062,
667
+ "step": 9964
668
+ },
669
+ {
670
+ "epoch": 48.0,
671
+ "learning_rate": 7.6e-05,
672
+ "loss": 0.0073,
673
+ "step": 10176
674
+ },
675
+ {
676
+ "epoch": 48.0,
677
+ "eval_loss": 0.5633374452590942,
678
+ "eval_runtime": 3.8494,
679
+ "eval_samples_per_second": 175.614,
680
+ "eval_steps_per_second": 22.082,
681
+ "step": 10176
682
+ },
683
+ {
684
+ "epoch": 49.0,
685
+ "learning_rate": 7.55e-05,
686
+ "loss": 0.0064,
687
+ "step": 10388
688
+ },
689
+ {
690
+ "epoch": 49.0,
691
+ "eval_loss": 0.558996319770813,
692
+ "eval_runtime": 3.8307,
693
+ "eval_samples_per_second": 176.467,
694
+ "eval_steps_per_second": 22.189,
695
+ "step": 10388
696
+ },
697
+ {
698
+ "epoch": 50.0,
699
+ "learning_rate": 7.500000000000001e-05,
700
+ "loss": 0.0047,
701
+ "step": 10600
702
+ },
703
+ {
704
+ "epoch": 50.0,
705
+ "eval_loss": 0.5562776923179626,
706
+ "eval_runtime": 3.8234,
707
+ "eval_samples_per_second": 176.804,
708
+ "eval_steps_per_second": 22.231,
709
+ "step": 10600
710
+ },
711
+ {
712
+ "epoch": 51.0,
713
+ "learning_rate": 7.450000000000001e-05,
714
+ "loss": 0.0107,
715
+ "step": 10812
716
+ },
717
+ {
718
+ "epoch": 51.0,
719
+ "eval_loss": 0.5780841708183289,
720
+ "eval_runtime": 3.8575,
721
+ "eval_samples_per_second": 175.242,
722
+ "eval_steps_per_second": 22.035,
723
+ "step": 10812
724
+ }
725
+ ],
726
+ "logging_steps": 500,
727
+ "max_steps": 42400,
728
+ "num_train_epochs": 200,
729
+ "save_steps": 500,
730
+ "total_flos": 1.3617372042962724e+16,
731
+ "trial_name": null,
732
+ "trial_params": null
733
+ }
checkpoint-10812/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8bbad506d5d7c9f6e6c369455b23a5312c048f6fdad38f9f3772905289326b4f
3
+ size 4155