loris3 commited on
Commit
b88fcee
·
verified ·
1 Parent(s): 5b1d271

Upload folder using huggingface_hub

Browse files
checkpoints/checkpoint-3547/config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "LlamaForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 1,
8
+ "eos_token_id": 2,
9
+ "head_dim": 64,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 768,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 2048,
14
+ "max_position_embeddings": 256,
15
+ "mlp_bias": false,
16
+ "model_type": "llama",
17
+ "num_attention_heads": 12,
18
+ "num_hidden_layers": 12,
19
+ "num_key_value_heads": 12,
20
+ "pad_token_id": 1,
21
+ "pretraining_tp": 1,
22
+ "rms_norm_eps": 1e-06,
23
+ "rope_scaling": null,
24
+ "rope_theta": 10000.0,
25
+ "tie_word_embeddings": true,
26
+ "torch_dtype": "float32",
27
+ "transformers_version": "4.47.0",
28
+ "use_cache": true,
29
+ "vocab_size": 16000
30
+ }
checkpoints/checkpoint-3547/generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "pad_token_id": 1,
6
+ "transformers_version": "4.47.0"
7
+ }
checkpoints/checkpoint-3547/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6219e4e6c96a2ea1fc357029f601be454176c20a48147ab45e7019bfba78a2e8
3
+ size 388979624
checkpoints/checkpoint-3547/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d6477fbd771f2014e6d2e52fba26514e8739f98e41ce5a0b80857346662fd922
3
+ size 778027770
checkpoints/checkpoint-3547/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c70a41264c08a1e8401b7173fe9901cfca41eb5cf987bd975ed722bcda9db818
3
+ size 14244
checkpoints/checkpoint-3547/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d1475294f68043bde4a2b641c96b3c189d4ec8ae3367f9dcd3ffdf6f70587948
3
+ size 1064
checkpoints/checkpoint-3547/trainer_state.json ADDED
@@ -0,0 +1,613 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 8.564458263573307,
5
+ "eval_steps": 500,
6
+ "global_step": 3547,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.09567089213106912,
13
+ "grad_norm": 19.851242065429688,
14
+ "learning_rate": 0.00011666666666666667,
15
+ "loss": 121.1525,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 0.19134178426213824,
20
+ "grad_norm": 13.892197608947754,
21
+ "learning_rate": 0.00023333333333333333,
22
+ "loss": 77.1477,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 0.2870126763932074,
27
+ "grad_norm": 12.60777759552002,
28
+ "learning_rate": 0.00035,
29
+ "loss": 64.8839,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 0.3826835685242765,
34
+ "grad_norm": 8.890196800231934,
35
+ "learning_rate": 0.00046666666666666666,
36
+ "loss": 59.8849,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 0.4783544606553456,
41
+ "grad_norm": 8.705379486083984,
42
+ "learning_rate": 0.0005833333333333334,
43
+ "loss": 57.5299,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 0.5740253527864148,
48
+ "grad_norm": 7.457199573516846,
49
+ "learning_rate": 0.0007,
50
+ "loss": 56.2806,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 0.6696962449174838,
55
+ "grad_norm": 7.943981647491455,
56
+ "learning_rate": 0.000699821634561209,
57
+ "loss": 55.3251,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 0.765367137048553,
62
+ "grad_norm": 7.716305732727051,
63
+ "learning_rate": 0.0006992867200404345,
64
+ "loss": 52.4636,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 0.8610380291796221,
69
+ "grad_norm": 7.123641014099121,
70
+ "learning_rate": 0.0006983958016391807,
71
+ "loss": 52.215,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 0.9567089213106912,
76
+ "grad_norm": 7.223691463470459,
77
+ "learning_rate": 0.0006971497874091708,
78
+ "loss": 52.1157,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 0.9988041138483617,
83
+ "eval_accuracy": 0.0,
84
+ "eval_loss": 8.116789817810059,
85
+ "eval_normalizer": 685885.0,
86
+ "eval_runtime": 103.9352,
87
+ "eval_samples_per_second": 514.33,
88
+ "eval_steps_per_second": 1.01,
89
+ "step": 522
90
+ },
91
+ {
92
+ "epoch": 1.0535756995933987,
93
+ "grad_norm": 7.065810203552246,
94
+ "learning_rate": 0.0006955499473268326,
95
+ "loss": 50.0351,
96
+ "step": 550
97
+ },
98
+ {
99
+ "epoch": 1.1492465917244679,
100
+ "grad_norm": 7.027311325073242,
101
+ "learning_rate": 0.0006935979119988993,
102
+ "loss": 48.6995,
103
+ "step": 600
104
+ },
105
+ {
106
+ "epoch": 1.244917483855537,
107
+ "grad_norm": 6.899056911468506,
108
+ "learning_rate": 0.0006912956710004438,
109
+ "loss": 48.9226,
110
+ "step": 650
111
+ },
112
+ {
113
+ "epoch": 1.3405883759866062,
114
+ "grad_norm": 7.2765116691589355,
115
+ "learning_rate": 0.0006886455708470427,
116
+ "loss": 48.9019,
117
+ "step": 700
118
+ },
119
+ {
120
+ "epoch": 1.4362592681176751,
121
+ "grad_norm": 8.293883323669434,
122
+ "learning_rate": 0.0006856503126031346,
123
+ "loss": 44.5473,
124
+ "step": 750
125
+ },
126
+ {
127
+ "epoch": 1.5319301602487443,
128
+ "grad_norm": 7.498640537261963,
129
+ "learning_rate": 0.0006823129491290102,
130
+ "loss": 45.238,
131
+ "step": 800
132
+ },
133
+ {
134
+ "epoch": 1.6276010523798135,
135
+ "grad_norm": 7.582442283630371,
136
+ "learning_rate": 0.0006786368819692442,
137
+ "loss": 45.6905,
138
+ "step": 850
139
+ },
140
+ {
141
+ "epoch": 1.7232719445108826,
142
+ "grad_norm": 8.016205787658691,
143
+ "learning_rate": 0.0006746258578857331,
144
+ "loss": 42.9074,
145
+ "step": 900
146
+ },
147
+ {
148
+ "epoch": 1.8189428366419516,
149
+ "grad_norm": 8.324728012084961,
150
+ "learning_rate": 0.000670283965038881,
151
+ "loss": 40.5995,
152
+ "step": 950
153
+ },
154
+ {
155
+ "epoch": 1.9146137287730207,
156
+ "grad_norm": 8.61531925201416,
157
+ "learning_rate": 0.0006656156288208179,
158
+ "loss": 41.5048,
159
+ "step": 1000
160
+ },
161
+ {
162
+ "epoch": 1.9988041138483617,
163
+ "eval_accuracy": 0.0,
164
+ "eval_loss": 8.638134956359863,
165
+ "eval_normalizer": 685885.0,
166
+ "eval_runtime": 104.5141,
167
+ "eval_samples_per_second": 511.481,
168
+ "eval_steps_per_second": 1.005,
169
+ "step": 1044
170
+ },
171
+ {
172
+ "epoch": 2.0114805070557282,
173
+ "grad_norm": 119.22761535644531,
174
+ "learning_rate": 0.000660625607344904,
175
+ "loss": 52.7863,
176
+ "step": 1050
177
+ },
178
+ {
179
+ "epoch": 2.1071513991867974,
180
+ "grad_norm": 7.394373893737793,
181
+ "learning_rate": 0.0006553189865961112,
182
+ "loss": 87.7065,
183
+ "step": 1100
184
+ },
185
+ {
186
+ "epoch": 2.2028222913178666,
187
+ "grad_norm": 5.8150200843811035,
188
+ "learning_rate": 0.0006497011752472301,
189
+ "loss": 68.8677,
190
+ "step": 1150
191
+ },
192
+ {
193
+ "epoch": 2.2984931834489357,
194
+ "grad_norm": 8.597038269042969,
195
+ "learning_rate": 0.0006437778991461825,
196
+ "loss": 59.723,
197
+ "step": 1200
198
+ },
199
+ {
200
+ "epoch": 2.3597225544128198,
201
+ "eval_accuracy": 0.0,
202
+ "eval_loss": 5.760004997253418,
203
+ "eval_normalizer": 685885.0,
204
+ "eval_runtime": 102.8572,
205
+ "eval_samples_per_second": 519.721,
206
+ "eval_steps_per_second": 1.021,
207
+ "step": 1232
208
+ },
209
+ {
210
+ "epoch": 3.0344415211671847,
211
+ "grad_norm": 9.546870231628418,
212
+ "learning_rate": 0.0006375551954800587,
213
+ "loss": 49.7983,
214
+ "step": 1250
215
+ },
216
+ {
217
+ "epoch": 3.130112413298254,
218
+ "grad_norm": 11.537946701049805,
219
+ "learning_rate": 0.0006310394066218296,
220
+ "loss": 38.324,
221
+ "step": 1300
222
+ },
223
+ {
224
+ "epoch": 3.225783305429323,
225
+ "grad_norm": 11.162705421447754,
226
+ "learning_rate": 0.0006242371736660025,
227
+ "loss": 27.47,
228
+ "step": 1350
229
+ },
230
+ {
231
+ "epoch": 3.321454197560392,
232
+ "grad_norm": 12.32886791229248,
233
+ "learning_rate": 0.000617155429659811,
234
+ "loss": 18.7372,
235
+ "step": 1400
236
+ },
237
+ {
238
+ "epoch": 3.3597225544128198,
239
+ "eval_accuracy": 0.0,
240
+ "eval_loss": 7.427099227905273,
241
+ "eval_normalizer": 685885.0,
242
+ "eval_runtime": 110.0904,
243
+ "eval_samples_per_second": 485.573,
244
+ "eval_steps_per_second": 0.954,
245
+ "step": 1420
246
+ },
247
+ {
248
+ "epoch": 4.057402535278642,
249
+ "grad_norm": 13.549798011779785,
250
+ "learning_rate": 0.0006098013925368385,
251
+ "loss": 57.3773,
252
+ "step": 1450
253
+ },
254
+ {
255
+ "epoch": 4.15307342740971,
256
+ "grad_norm": 10.063131332397461,
257
+ "learning_rate": 0.0006021825577602754,
258
+ "loss": 71.4705,
259
+ "step": 1500
260
+ },
261
+ {
262
+ "epoch": 4.24874431954078,
263
+ "grad_norm": 9.314085006713867,
264
+ "learning_rate": 0.0005943066906833104,
265
+ "loss": 69.1017,
266
+ "step": 1550
267
+ },
268
+ {
269
+ "epoch": 4.344415211671849,
270
+ "grad_norm": 8.48709774017334,
271
+ "learning_rate": 0.0005861818186344407,
272
+ "loss": 66.7627,
273
+ "step": 1600
274
+ },
275
+ {
276
+ "epoch": 4.440086103802918,
277
+ "grad_norm": 9.457844734191895,
278
+ "learning_rate": 0.00057781622273577,
279
+ "loss": 62.5878,
280
+ "step": 1650
281
+ },
282
+ {
283
+ "epoch": 4.535756995933987,
284
+ "grad_norm": 8.216409683227539,
285
+ "learning_rate": 0.0005692184294626307,
286
+ "loss": 63.0275,
287
+ "step": 1700
288
+ },
289
+ {
290
+ "epoch": 4.631427888065057,
291
+ "grad_norm": 8.80875301361084,
292
+ "learning_rate": 0.0005603972019531362,
293
+ "loss": 62.8742,
294
+ "step": 1750
295
+ },
296
+ {
297
+ "epoch": 4.727098780196125,
298
+ "grad_norm": 9.469932556152344,
299
+ "learning_rate": 0.0005513615310765172,
300
+ "loss": 56.6002,
301
+ "step": 1800
302
+ },
303
+ {
304
+ "epoch": 4.822769672327194,
305
+ "grad_norm": 9.208184242248535,
306
+ "learning_rate": 0.0005421206262693491,
307
+ "loss": 55.5842,
308
+ "step": 1850
309
+ },
310
+ {
311
+ "epoch": 4.918440564458264,
312
+ "grad_norm": 9.542802810668945,
313
+ "learning_rate": 0.0005326839061490078,
314
+ "loss": 56.4509,
315
+ "step": 1900
316
+ },
317
+ {
318
+ "epoch": 4.998804113848362,
319
+ "eval_accuracy": 0.0,
320
+ "eval_loss": 5.560527801513672,
321
+ "eval_normalizer": 685885.0,
322
+ "eval_runtime": 102.9429,
323
+ "eval_samples_per_second": 519.288,
324
+ "eval_steps_per_second": 1.02,
325
+ "step": 1942
326
+ },
327
+ {
328
+ "epoch": 5.015307342740971,
329
+ "grad_norm": 10.22252082824707,
330
+ "learning_rate": 0.0005230609889139216,
331
+ "loss": 52.8616,
332
+ "step": 1950
333
+ },
334
+ {
335
+ "epoch": 5.11097823487204,
336
+ "grad_norm": 11.31312370300293,
337
+ "learning_rate": 0.0005132616825404055,
338
+ "loss": 45.5469,
339
+ "step": 2000
340
+ },
341
+ {
342
+ "epoch": 5.206649127003109,
343
+ "grad_norm": 11.319524765014648,
344
+ "learning_rate": 0.0005032959747860662,
345
+ "loss": 47.203,
346
+ "step": 2050
347
+ },
348
+ {
349
+ "epoch": 5.302320019134179,
350
+ "grad_norm": 11.748647689819336,
351
+ "learning_rate": 0.000493174023009969,
352
+ "loss": 48.3289,
353
+ "step": 2100
354
+ },
355
+ {
356
+ "epoch": 5.397990911265247,
357
+ "grad_norm": 12.092212677001953,
358
+ "learning_rate": 0.00048290614381994235,
359
+ "loss": 39.7512,
360
+ "step": 2150
361
+ },
362
+ {
363
+ "epoch": 5.493661803396317,
364
+ "grad_norm": 12.723077774047852,
365
+ "learning_rate": 0.00047250280255757023,
366
+ "loss": 38.0036,
367
+ "step": 2200
368
+ },
369
+ {
370
+ "epoch": 5.589332695527386,
371
+ "grad_norm": 13.034537315368652,
372
+ "learning_rate": 0.0004619746026315906,
373
+ "loss": 39.4596,
374
+ "step": 2250
375
+ },
376
+ {
377
+ "epoch": 5.685003587658455,
378
+ "grad_norm": 11.695505142211914,
379
+ "learning_rate": 0.00045133227471057203,
380
+ "loss": 36.7652,
381
+ "step": 2300
382
+ },
383
+ {
384
+ "epoch": 5.780674479789524,
385
+ "grad_norm": 12.36651611328125,
386
+ "learning_rate": 0.00044058666578588224,
387
+ "loss": 30.5917,
388
+ "step": 2350
389
+ },
390
+ {
391
+ "epoch": 5.876345371920593,
392
+ "grad_norm": 12.983572006225586,
393
+ "learning_rate": 0.0004297487281160982,
394
+ "loss": 32.1428,
395
+ "step": 2400
396
+ },
397
+ {
398
+ "epoch": 5.972016264051662,
399
+ "grad_norm": 13.80376148223877,
400
+ "learning_rate": 0.00041882950806412285,
401
+ "loss": 33.0843,
402
+ "step": 2450
403
+ },
404
+ {
405
+ "epoch": 5.998804113848362,
406
+ "eval_accuracy": 0.0,
407
+ "eval_loss": 6.923346042633057,
408
+ "eval_normalizer": 685885.0,
409
+ "eval_runtime": 103.7481,
410
+ "eval_samples_per_second": 515.258,
411
+ "eval_steps_per_second": 1.012,
412
+ "step": 2464
413
+ },
414
+ {
415
+ "epoch": 6.068883042334369,
416
+ "grad_norm": 9.631818771362305,
417
+ "learning_rate": 0.0004078401348383897,
418
+ "loss": 65.9342,
419
+ "step": 2500
420
+ },
421
+ {
422
+ "epoch": 6.164553934465439,
423
+ "grad_norm": 7.13682746887207,
424
+ "learning_rate": 0.00039679180914962693,
425
+ "loss": 70.9535,
426
+ "step": 2550
427
+ },
428
+ {
429
+ "epoch": 6.260224826596508,
430
+ "grad_norm": 7.001780033111572,
431
+ "learning_rate": 0.00038569579179474536,
432
+ "loss": 65.2834,
433
+ "step": 2600
434
+ },
435
+ {
436
+ "epoch": 6.355895718727577,
437
+ "grad_norm": 8.191680908203125,
438
+ "learning_rate": 0.00037456339217948394,
439
+ "loss": 60.6176,
440
+ "step": 2650
441
+ },
442
+ {
443
+ "epoch": 6.451566610858646,
444
+ "grad_norm": 8.527460098266602,
445
+ "learning_rate": 0.0003634059567915124,
446
+ "loss": 57.9774,
447
+ "step": 2700
448
+ },
449
+ {
450
+ "epoch": 6.547237502989716,
451
+ "grad_norm": 10.255255699157715,
452
+ "learning_rate": 0.00035223485763573775,
453
+ "loss": 48.2276,
454
+ "step": 2750
455
+ },
456
+ {
457
+ "epoch": 6.642908395120784,
458
+ "grad_norm": 11.824259757995605,
459
+ "learning_rate": 0.00034106148064360405,
460
+ "loss": 44.7545,
461
+ "step": 2800
462
+ },
463
+ {
464
+ "epoch": 6.738579287251854,
465
+ "grad_norm": 12.938841819763184,
466
+ "learning_rate": 0.0003298972140681969,
467
+ "loss": 39.2877,
468
+ "step": 2850
469
+ },
470
+ {
471
+ "epoch": 6.753886629992825,
472
+ "eval_accuracy": 0.0,
473
+ "eval_loss": 5.779562950134277,
474
+ "eval_normalizer": 685885.0,
475
+ "eval_runtime": 104.2115,
476
+ "eval_samples_per_second": 512.967,
477
+ "eval_steps_per_second": 1.008,
478
+ "step": 2858
479
+ },
480
+ {
481
+ "epoch": 7.080363549390098,
482
+ "grad_norm": 13.708643913269043,
483
+ "learning_rate": 0.00031875343687698203,
484
+ "loss": 31.5294,
485
+ "step": 2900
486
+ },
487
+ {
488
+ "epoch": 7.176034441521167,
489
+ "grad_norm": 13.05090618133545,
490
+ "learning_rate": 0.000307641507154008,
491
+ "loss": 28.8423,
492
+ "step": 2950
493
+ },
494
+ {
495
+ "epoch": 7.271705333652236,
496
+ "grad_norm": 15.017513275146484,
497
+ "learning_rate": 0.0002965727505233939,
498
+ "loss": 23.1749,
499
+ "step": 3000
500
+ },
501
+ {
502
+ "epoch": 7.367376225783305,
503
+ "grad_norm": 13.55999755859375,
504
+ "learning_rate": 0.0002855584486059016,
505
+ "loss": 19.6325,
506
+ "step": 3050
507
+ },
508
+ {
509
+ "epoch": 7.463047117914375,
510
+ "grad_norm": 10.594040870666504,
511
+ "learning_rate": 0.00027460982752035653,
512
+ "loss": 18.3422,
513
+ "step": 3100
514
+ },
515
+ {
516
+ "epoch": 7.558718010045443,
517
+ "grad_norm": 13.867926597595215,
518
+ "learning_rate": 0.00026373804644163934,
519
+ "loss": 14.5862,
520
+ "step": 3150
521
+ },
522
+ {
523
+ "epoch": 7.654388902176513,
524
+ "grad_norm": 11.124190330505371,
525
+ "learning_rate": 0.00025295418622690716,
526
+ "loss": 13.728,
527
+ "step": 3200
528
+ },
529
+ {
530
+ "epoch": 7.750059794307582,
531
+ "grad_norm": 13.195384979248047,
532
+ "learning_rate": 0.00024226923812164069,
533
+ "loss": 13.0752,
534
+ "step": 3250
535
+ },
536
+ {
537
+ "epoch": 7.753886629992825,
538
+ "eval_accuracy": 0.0,
539
+ "eval_loss": 7.642109394073486,
540
+ "eval_normalizer": 685885.0,
541
+ "eval_runtime": 103.2383,
542
+ "eval_samples_per_second": 517.802,
543
+ "eval_steps_per_second": 1.017,
544
+ "step": 3252
545
+ },
546
+ {
547
+ "epoch": 8.091844056445826,
548
+ "grad_norm": 7.070100784301758,
549
+ "learning_rate": 0.0002316940925570241,
550
+ "loss": 83.8433,
551
+ "step": 3300
552
+ },
553
+ {
554
+ "epoch": 8.187514948576895,
555
+ "grad_norm": 7.2077250480651855,
556
+ "learning_rate": 0.0002212395280500796,
557
+ "loss": 73.3812,
558
+ "step": 3350
559
+ },
560
+ {
561
+ "epoch": 8.283185840707965,
562
+ "grad_norm": 8.034393310546875,
563
+ "learning_rate": 0.00021091620021786818,
564
+ "loss": 68.4622,
565
+ "step": 3400
566
+ },
567
+ {
568
+ "epoch": 8.378856732839033,
569
+ "grad_norm": 9.863322257995605,
570
+ "learning_rate": 0.0002007346309169531,
571
+ "loss": 63.8324,
572
+ "step": 3450
573
+ },
574
+ {
575
+ "epoch": 8.474527624970102,
576
+ "grad_norm": 11.584691047668457,
577
+ "learning_rate": 0.00019070519751919701,
578
+ "loss": 58.9669,
579
+ "step": 3500
580
+ },
581
+ {
582
+ "epoch": 8.564458263573307,
583
+ "eval_accuracy": 0.0,
584
+ "eval_loss": 5.5344061851501465,
585
+ "eval_normalizer": 685885.0,
586
+ "eval_runtime": 103.9205,
587
+ "eval_samples_per_second": 514.403,
588
+ "eval_steps_per_second": 1.01,
589
+ "step": 3547
590
+ }
591
+ ],
592
+ "logging_steps": 50,
593
+ "max_steps": 5220,
594
+ "num_input_tokens_seen": 0,
595
+ "num_train_epochs": 10,
596
+ "save_steps": 500,
597
+ "stateful_callbacks": {
598
+ "TrainerControl": {
599
+ "args": {
600
+ "should_epoch_stop": false,
601
+ "should_evaluate": false,
602
+ "should_log": false,
603
+ "should_save": true,
604
+ "should_training_stop": false
605
+ },
606
+ "attributes": {}
607
+ }
608
+ },
609
+ "total_flos": 3.934216551069942e+17,
610
+ "train_batch_size": 128,
611
+ "trial_name": null,
612
+ "trial_params": null
613
+ }
checkpoints/checkpoint-3547/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:112439a8bf818313a9112c0345790e09c2dd4279f702a6944e87378fe04ce172
3
+ size 5432