ChiefTheLord commited on
Commit
abfe160
·
verified ·
1 Parent(s): d56b9c3

Upload folder using huggingface_hub

Browse files
checkpoints-v5.0-discrete/checkpoint-20480/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc743da0f60e65d4074b36b0e16ed6c0941b61bb6bd6d796a401711b54d1fb39
3
+ size 24033312
checkpoints-v5.0-discrete/checkpoint-20480/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:758585e59d739d4567df00dde3308a77f0ce62305b8f10910493b453cd9ddc70
3
+ size 1641995
checkpoints-v5.0-discrete/checkpoint-20480/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:acf95e4d18ad058ca5e9d2d3000a021bfed64b42ba15a106f9bc7f890794c620
3
+ size 14645
checkpoints-v5.0-discrete/checkpoint-20480/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:74273fde28f4d2d8661f02c97a3fdc5a64c59bc3341b1e39cc74b538cd469ac2
3
+ size 1383
checkpoints-v5.0-discrete/checkpoint-20480/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:205d765e13e1be2966644dd0f0eaaf87a8a81b7f8860b6a9dc2ebf25b30616e3
3
+ size 1465
checkpoints-v5.0-discrete/checkpoint-20480/trainer_state.json ADDED
@@ -0,0 +1,679 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.945914738349268,
6
+ "eval_steps": 4096,
7
+ "global_step": 20480,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.011823934229365849,
14
+ "grad_norm": 2.0670883655548096,
15
+ "learning_rate": 0.000498046875,
16
+ "loss": 1.120773196220398,
17
+ "step": 256
18
+ },
19
+ {
20
+ "epoch": 0.023647868458731697,
21
+ "grad_norm": 2.177950382232666,
22
+ "learning_rate": 0.000998046875,
23
+ "loss": 1.1360918283462524,
24
+ "step": 512
25
+ },
26
+ {
27
+ "epoch": 0.03547180268809755,
28
+ "grad_norm": 1.7646818161010742,
29
+ "learning_rate": 0.000999640996023194,
30
+ "loss": 1.1378288269042969,
31
+ "step": 768
32
+ },
33
+ {
34
+ "epoch": 0.047295736917463395,
35
+ "grad_norm": 2.6680169105529785,
36
+ "learning_rate": 0.0009985588674043958,
37
+ "loss": 1.135412335395813,
38
+ "step": 1024
39
+ },
40
+ {
41
+ "epoch": 0.05911967114682925,
42
+ "grad_norm": 1.831610083580017,
43
+ "learning_rate": 0.0009967551747861387,
44
+ "loss": 1.137319564819336,
45
+ "step": 1280
46
+ },
47
+ {
48
+ "epoch": 0.0709436053761951,
49
+ "grad_norm": 2.298506736755371,
50
+ "learning_rate": 0.000994232528651847,
51
+ "loss": 1.1369880437850952,
52
+ "step": 1536
53
+ },
54
+ {
55
+ "epoch": 0.08276753960556095,
56
+ "grad_norm": 2.3097012042999268,
57
+ "learning_rate": 0.0009909945800260092,
58
+ "loss": 1.1359608173370361,
59
+ "step": 1792
60
+ },
61
+ {
62
+ "epoch": 0.09459147383492679,
63
+ "grad_norm": 1.7380794286727905,
64
+ "learning_rate": 0.0009870460151900522,
65
+ "loss": 1.1206920146942139,
66
+ "step": 2048
67
+ },
68
+ {
69
+ "epoch": 0.10641540806429264,
70
+ "grad_norm": 1.5254817008972168,
71
+ "learning_rate": 0.0009823925488998885,
72
+ "loss": 1.1205227375030518,
73
+ "step": 2304
74
+ },
75
+ {
76
+ "epoch": 0.1182393422936585,
77
+ "grad_norm": 1.821317195892334,
78
+ "learning_rate": 0.0009770409161149525,
79
+ "loss": 1.1351104974746704,
80
+ "step": 2560
81
+ },
82
+ {
83
+ "epoch": 0.13006327652302435,
84
+ "grad_norm": 1.91286039352417,
85
+ "learning_rate": 0.0009709988622506973,
86
+ "loss": 1.1217825412750244,
87
+ "step": 2816
88
+ },
89
+ {
90
+ "epoch": 0.1418872107523902,
91
+ "grad_norm": 1.878923773765564,
92
+ "learning_rate": 0.000964275131968659,
93
+ "loss": 1.122628927230835,
94
+ "step": 3072
95
+ },
96
+ {
97
+ "epoch": 0.15371114498175603,
98
+ "grad_norm": 2.3001058101654053,
99
+ "learning_rate": 0.0009568794565203123,
100
+ "loss": 1.111568808555603,
101
+ "step": 3328
102
+ },
103
+ {
104
+ "epoch": 0.1655350792111219,
105
+ "grad_norm": 2.146392583847046,
106
+ "learning_rate": 0.0009488225396630347,
107
+ "loss": 1.1250687837600708,
108
+ "step": 3584
109
+ },
110
+ {
111
+ "epoch": 0.17735901344048774,
112
+ "grad_norm": 2.3365514278411865,
113
+ "learning_rate": 0.0009401160421685646,
114
+ "loss": 1.122818946838379,
115
+ "step": 3840
116
+ },
117
+ {
118
+ "epoch": 0.18918294766985358,
119
+ "grad_norm": 1.1318840980529785,
120
+ "learning_rate": 0.0009307725649463714,
121
+ "loss": 1.1216130256652832,
122
+ "step": 4096
123
+ },
124
+ {
125
+ "epoch": 0.18918294766985358,
126
+ "eval_cos_loss": 0.26913025553368,
127
+ "eval_loss": 1.1185425997596898,
128
+ "eval_mse_loss": 1.091629574696223,
129
+ "step": 4096
130
+ },
131
+ {
132
+ "epoch": 0.18918294766985358,
133
+ "eval_cos_loss": 0.26913025553368,
134
+ "eval_loss": 1.1185425997596898,
135
+ "eval_mse_loss": 1.091629574696223,
136
+ "eval_runtime": 28.2356,
137
+ "eval_samples_per_second": 991.409,
138
+ "eval_steps_per_second": 15.512,
139
+ "step": 4096
140
+ },
141
+ {
142
+ "epoch": 0.20100688189921945,
143
+ "grad_norm": 2.2528905868530273,
144
+ "learning_rate": 0.0009208056308063659,
145
+ "loss": 1.1088756322860718,
146
+ "step": 4352
147
+ },
148
+ {
149
+ "epoch": 0.2128308161285853,
150
+ "grad_norm": 2.0888235569000244,
151
+ "learning_rate": 0.0009102296648873445,
152
+ "loss": 1.127668023109436,
153
+ "step": 4608
154
+ },
155
+ {
156
+ "epoch": 0.22465475035795113,
157
+ "grad_norm": 2.7774133682250977,
158
+ "learning_rate": 0.0008990599737794927,
159
+ "loss": 1.124489426612854,
160
+ "step": 4864
161
+ },
162
+ {
163
+ "epoch": 0.236478684587317,
164
+ "grad_norm": 1.200258731842041,
165
+ "learning_rate": 0.0008873127233711644,
166
+ "loss": 1.1282005310058594,
167
+ "step": 5120
168
+ },
169
+ {
170
+ "epoch": 0.24830261881668284,
171
+ "grad_norm": 1.4780941009521484,
172
+ "learning_rate": 0.0008750049154520011,
173
+ "loss": 1.1131128072738647,
174
+ "step": 5376
175
+ },
176
+ {
177
+ "epoch": 0.2601265530460487,
178
+ "grad_norm": 1.460057258605957,
179
+ "learning_rate": 0.0008621543631062487,
180
+ "loss": 1.108135461807251,
181
+ "step": 5632
182
+ },
183
+ {
184
+ "epoch": 0.27195048727541454,
185
+ "grad_norm": 2.2862741947174072,
186
+ "learning_rate": 0.0008487796649318904,
187
+ "loss": 1.1027661561965942,
188
+ "step": 5888
189
+ },
190
+ {
191
+ "epoch": 0.2837744215047804,
192
+ "grad_norm": 2.866795539855957,
193
+ "learning_rate": 0.0008349001781229053,
194
+ "loss": 1.110382080078125,
195
+ "step": 6144
196
+ },
197
+ {
198
+ "epoch": 0.2955983557341462,
199
+ "grad_norm": 1.7982397079467773,
200
+ "learning_rate": 0.0008205359904536107,
201
+ "loss": 1.0966479778289795,
202
+ "step": 6400
203
+ },
204
+ {
205
+ "epoch": 0.30742228996351206,
206
+ "grad_norm": 2.270012855529785,
207
+ "learning_rate": 0.0008057078912056363,
208
+ "loss": 1.111327886581421,
209
+ "step": 6656
210
+ },
211
+ {
212
+ "epoch": 0.3192462241928779,
213
+ "grad_norm": 2.4083364009857178,
214
+ "learning_rate": 0.0007904373410796086,
215
+ "loss": 1.0966944694519043,
216
+ "step": 6912
217
+ },
218
+ {
219
+ "epoch": 0.3310701584222438,
220
+ "grad_norm": 1.2960336208343506,
221
+ "learning_rate": 0.0007747464411350876,
222
+ "loss": 1.101299524307251,
223
+ "step": 7168
224
+ },
225
+ {
226
+ "epoch": 0.34289409265160964,
227
+ "grad_norm": 1.9119939804077148,
228
+ "learning_rate": 0.000758657900803716,
229
+ "loss": 1.1009973287582397,
230
+ "step": 7424
231
+ },
232
+ {
233
+ "epoch": 0.3547180268809755,
234
+ "grad_norm": 1.7134813070297241,
235
+ "learning_rate": 0.000742195005021869,
236
+ "loss": 1.0911792516708374,
237
+ "step": 7680
238
+ },
239
+ {
240
+ "epoch": 0.3665419611103413,
241
+ "grad_norm": 0.9328792691230774,
242
+ "learning_rate": 0.0007253815805303786,
243
+ "loss": 1.1016789674758911,
244
+ "step": 7936
245
+ },
246
+ {
247
+ "epoch": 0.37836589533970716,
248
+ "grad_norm": 1.0006275177001953,
249
+ "learning_rate": 0.0007082419613901028,
250
+ "loss": 1.1058346033096313,
251
+ "step": 8192
252
+ },
253
+ {
254
+ "epoch": 0.37836589533970716,
255
+ "eval_cos_loss": 0.26391498703662664,
256
+ "eval_loss": 1.095669688427285,
257
+ "eval_mse_loss": 1.0692781910776548,
258
+ "step": 8192
259
+ },
260
+ {
261
+ "epoch": 0.37836589533970716,
262
+ "eval_cos_loss": 0.26391498703662664,
263
+ "eval_loss": 1.095669688427285,
264
+ "eval_mse_loss": 1.0692781910776548,
265
+ "eval_runtime": 26.9937,
266
+ "eval_samples_per_second": 1037.02,
267
+ "eval_steps_per_second": 16.226,
268
+ "step": 8192
269
+ },
270
+ {
271
+ "epoch": 0.390189829569073,
272
+ "grad_norm": 1.500245213508606,
273
+ "learning_rate": 0.0006908009537632514,
274
+ "loss": 1.0995627641677856,
275
+ "step": 8448
276
+ },
277
+ {
278
+ "epoch": 0.4020137637984389,
279
+ "grad_norm": 2.41027569770813,
280
+ "learning_rate": 0.0006730838000114403,
281
+ "loss": 1.1009161472320557,
282
+ "step": 8704
283
+ },
284
+ {
285
+ "epoch": 0.41383769802780473,
286
+ "grad_norm": 1.3940221071243286,
287
+ "learning_rate": 0.0006551161421624341,
288
+ "loss": 1.1009058952331543,
289
+ "step": 8960
290
+ },
291
+ {
292
+ "epoch": 0.4256616322571706,
293
+ "grad_norm": 1.4039406776428223,
294
+ "learning_rate": 0.0006369239847984517,
295
+ "loss": 1.0966987609863281,
296
+ "step": 9216
297
+ },
298
+ {
299
+ "epoch": 0.4374855664865364,
300
+ "grad_norm": 2.5753512382507324,
301
+ "learning_rate": 0.0006185336574197479,
302
+ "loss": 1.0883150100708008,
303
+ "step": 9472
304
+ },
305
+ {
306
+ "epoch": 0.44930950071590225,
307
+ "grad_norm": 0.6999452114105225,
308
+ "learning_rate": 0.0005999717763379407,
309
+ "loss": 1.0972821712493896,
310
+ "step": 9728
311
+ },
312
+ {
313
+ "epoch": 0.4611334349452681,
314
+ "grad_norm": 1.2475601434707642,
315
+ "learning_rate": 0.0005812652061542363,
316
+ "loss": 1.096147060394287,
317
+ "step": 9984
318
+ },
319
+ {
320
+ "epoch": 0.472957369174634,
321
+ "grad_norm": 1.8508833646774292,
322
+ "learning_rate": 0.0005624410208783071,
323
+ "loss": 1.0926183462142944,
324
+ "step": 10240
325
+ },
326
+ {
327
+ "epoch": 0.48478130340399983,
328
+ "grad_norm": 2.726325273513794,
329
+ "learning_rate": 0.0005435264647440881,
330
+ "loss": 1.0811611413955688,
331
+ "step": 10496
332
+ },
333
+ {
334
+ "epoch": 0.49660523763336567,
335
+ "grad_norm": 1.286672830581665,
336
+ "learning_rate": 0.000524548912779213,
337
+ "loss": 1.091227650642395,
338
+ "step": 10752
339
+ },
340
+ {
341
+ "epoch": 0.5084291718627315,
342
+ "grad_norm": 1.2610856294631958,
343
+ "learning_rate": 0.0005055358311851499,
344
+ "loss": 1.0942208766937256,
345
+ "step": 11008
346
+ },
347
+ {
348
+ "epoch": 0.5202531060920974,
349
+ "grad_norm": 1.3045330047607422,
350
+ "learning_rate": 0.0004865147375853812,
351
+ "loss": 1.082557201385498,
352
+ "step": 11264
353
+ },
354
+ {
355
+ "epoch": 0.5320770403214632,
356
+ "grad_norm": 1.2854019403457642,
357
+ "learning_rate": 0.0004675131611991607,
358
+ "loss": 1.0753118991851807,
359
+ "step": 11520
360
+ },
361
+ {
362
+ "epoch": 0.5439009745508291,
363
+ "grad_norm": 1.157984733581543,
364
+ "learning_rate": 0.0004485586029984899,
365
+ "loss": 1.092992901802063,
366
+ "step": 11776
367
+ },
368
+ {
369
+ "epoch": 0.5557249087801949,
370
+ "grad_norm": 1.6725375652313232,
371
+ "learning_rate": 0.00042967849590597266,
372
+ "loss": 1.084574580192566,
373
+ "step": 12032
374
+ },
375
+ {
376
+ "epoch": 0.5675488430095608,
377
+ "grad_norm": 1.9504145383834839,
378
+ "learning_rate": 0.0004109001650911621,
379
+ "loss": 1.0776673555374146,
380
+ "step": 12288
381
+ },
382
+ {
383
+ "epoch": 0.5675488430095608,
384
+ "eval_cos_loss": 0.25928048646613344,
385
+ "eval_loss": 1.0722966102976776,
386
+ "eval_mse_loss": 1.0463685593376422,
387
+ "step": 12288
388
+ },
389
+ {
390
+ "epoch": 0.5675488430095608,
391
+ "eval_cos_loss": 0.25928048646613344,
392
+ "eval_loss": 1.0722966102976776,
393
+ "eval_mse_loss": 1.0463685593376422,
394
+ "eval_runtime": 27.2665,
395
+ "eval_samples_per_second": 1026.643,
396
+ "eval_steps_per_second": 16.064,
397
+ "step": 12288
398
+ },
399
+ {
400
+ "epoch": 0.5793727772389267,
401
+ "grad_norm": 2.214651107788086,
402
+ "learning_rate": 0.0003922507884228551,
403
+ "loss": 1.0699119567871094,
404
+ "step": 12544
405
+ },
406
+ {
407
+ "epoch": 0.5911967114682924,
408
+ "grad_norm": 2.347487449645996,
409
+ "learning_rate": 0.00037375735713457723,
410
+ "loss": 1.0712252855300903,
411
+ "step": 12800
412
+ },
413
+ {
414
+ "epoch": 0.6030206456976583,
415
+ "grad_norm": 1.2330212593078613,
416
+ "learning_rate": 0.00035544663676018276,
417
+ "loss": 1.0869075059890747,
418
+ "step": 13056
419
+ },
420
+ {
421
+ "epoch": 0.6148445799270241,
422
+ "grad_norm": 1.9649189710617065,
423
+ "learning_rate": 0.00033734512839611255,
424
+ "loss": 1.079498291015625,
425
+ "step": 13312
426
+ },
427
+ {
428
+ "epoch": 0.62666851415639,
429
+ "grad_norm": 1.8303130865097046,
430
+ "learning_rate": 0.0003194790303463687,
431
+ "loss": 1.077008843421936,
432
+ "step": 13568
433
+ },
434
+ {
435
+ "epoch": 0.6384924483857558,
436
+ "grad_norm": 2.8677024841308594,
437
+ "learning_rate": 0.00030187420020572406,
438
+ "loss": 1.0742326974868774,
439
+ "step": 13824
440
+ },
441
+ {
442
+ "epoch": 0.6503163826151217,
443
+ "grad_norm": 1.6343201398849487,
444
+ "learning_rate": 0.00028455611743603626,
445
+ "loss": 1.0788600444793701,
446
+ "step": 14080
447
+ },
448
+ {
449
+ "epoch": 0.6621403168444876,
450
+ "grad_norm": 1.9170905351638794,
451
+ "learning_rate": 0.0002675498464898373,
452
+ "loss": 1.0711390972137451,
453
+ "step": 14336
454
+ },
455
+ {
456
+ "epoch": 0.6739642510738534,
457
+ "grad_norm": 2.6139180660247803,
458
+ "learning_rate": 0.0002508800005345623,
459
+ "loss": 1.0705795288085938,
460
+ "step": 14592
461
+ },
462
+ {
463
+ "epoch": 0.6857881853032193,
464
+ "grad_norm": 2.0574164390563965,
465
+ "learning_rate": 0.00023457070582992562,
466
+ "loss": 1.0699050426483154,
467
+ "step": 14848
468
+ },
469
+ {
470
+ "epoch": 0.6976121195325851,
471
+ "grad_norm": 1.8634923696517944,
472
+ "learning_rate": 0.00021864556680999692,
473
+ "loss": 1.0736559629440308,
474
+ "step": 15104
475
+ },
476
+ {
477
+ "epoch": 0.709436053761951,
478
+ "grad_norm": 1.8115946054458618,
479
+ "learning_rate": 0.0002031276319205152,
480
+ "loss": 1.0778332948684692,
481
+ "step": 15360
482
+ },
483
+ {
484
+ "epoch": 0.7212599879913169,
485
+ "grad_norm": 1.888077735900879,
486
+ "learning_rate": 0.00018803936026088542,
487
+ "loss": 1.0760120153427124,
488
+ "step": 15616
489
+ },
490
+ {
491
+ "epoch": 0.7330839222206826,
492
+ "grad_norm": 1.6118661165237427,
493
+ "learning_rate": 0.00017340258907913464,
494
+ "loss": 1.0667051076889038,
495
+ "step": 15872
496
+ },
497
+ {
498
+ "epoch": 0.7449078564500485,
499
+ "grad_norm": 0.7959473729133606,
500
+ "learning_rate": 0.0001592385021668743,
501
+ "loss": 1.0717016458511353,
502
+ "step": 16128
503
+ },
504
+ {
505
+ "epoch": 0.7567317906794143,
506
+ "grad_norm": 1.1920806169509888,
507
+ "learning_rate": 0.0001455675992000087,
508
+ "loss": 1.0653103590011597,
509
+ "step": 16384
510
+ },
511
+ {
512
+ "epoch": 0.7567317906794143,
513
+ "eval_cos_loss": 0.26112998185092456,
514
+ "eval_loss": 1.0791749034298066,
515
+ "eval_mse_loss": 1.0530619049725467,
516
+ "step": 16384
517
+ },
518
+ {
519
+ "epoch": 0.7567317906794143,
520
+ "eval_cos_loss": 0.26112998185092456,
521
+ "eval_loss": 1.0791749034298066,
522
+ "eval_mse_loss": 1.0530619049725467,
523
+ "eval_runtime": 26.307,
524
+ "eval_samples_per_second": 1064.089,
525
+ "eval_steps_per_second": 16.65,
526
+ "step": 16384
527
+ },
528
+ {
529
+ "epoch": 0.7685557249087802,
530
+ "grad_norm": 1.9962712526321411,
531
+ "learning_rate": 0.000132409666069565,
532
+ "loss": 1.0701987743377686,
533
+ "step": 16640
534
+ },
535
+ {
536
+ "epoch": 0.780379659138146,
537
+ "grad_norm": 0.9719748497009277,
538
+ "learning_rate": 0.0001197837462455823,
539
+ "loss": 1.060810923576355,
540
+ "step": 16896
541
+ },
542
+ {
543
+ "epoch": 0.7922035933675119,
544
+ "grad_norm": 1.1240458488464355,
545
+ "learning_rate": 0.00010770811321550749,
546
+ "loss": 1.0603857040405273,
547
+ "step": 17152
548
+ },
549
+ {
550
+ "epoch": 0.8040275275968778,
551
+ "grad_norm": 2.0561318397521973,
552
+ "learning_rate": 9.620024403698591e-05,
553
+ "loss": 1.0669511556625366,
554
+ "step": 17408
555
+ },
556
+ {
557
+ "epoch": 0.8158514618262436,
558
+ "grad_norm": 1.8854562044143677,
559
+ "learning_rate": 8.527679404332429e-05,
560
+ "loss": 1.0674059391021729,
561
+ "step": 17664
562
+ },
563
+ {
564
+ "epoch": 0.8276753960556095,
565
+ "grad_norm": 1.6353553533554077,
566
+ "learning_rate": 7.495357273823544e-05,
567
+ "loss": 1.0606721639633179,
568
+ "step": 17920
569
+ },
570
+ {
571
+ "epoch": 0.8394993302849753,
572
+ "grad_norm": 1.6140296459197998,
573
+ "learning_rate": 6.524552091475183e-05,
574
+ "loss": 1.0758994817733765,
575
+ "step": 18176
576
+ },
577
+ {
578
+ "epoch": 0.8513232645143411,
579
+ "grad_norm": 1.5846794843673706,
580
+ "learning_rate": 5.6166689031422024e-05,
581
+ "loss": 1.0615742206573486,
582
+ "step": 18432
583
+ },
584
+ {
585
+ "epoch": 0.8631471987437069,
586
+ "grad_norm": 2.6828203201293945,
587
+ "learning_rate": 4.773021687709067e-05,
588
+ "loss": 1.0637918710708618,
589
+ "step": 18688
590
+ },
591
+ {
592
+ "epoch": 0.8749711329730728,
593
+ "grad_norm": 2.613208770751953,
594
+ "learning_rate": 3.994831455368719e-05,
595
+ "loss": 1.063633680343628,
596
+ "step": 18944
597
+ },
598
+ {
599
+ "epoch": 0.8867950672024387,
600
+ "grad_norm": 1.3246995210647583,
601
+ "learning_rate": 3.283224480455282e-05,
602
+ "loss": 1.065570592880249,
603
+ "step": 19200
604
+ },
605
+ {
606
+ "epoch": 0.8986190014318045,
607
+ "grad_norm": 1.740508794784546,
608
+ "learning_rate": 2.639230671387627e-05,
609
+ "loss": 1.0593317747116089,
610
+ "step": 19456
611
+ },
612
+ {
613
+ "epoch": 0.9104429356611704,
614
+ "grad_norm": 1.9934636354446411,
615
+ "learning_rate": 2.063782080083576e-05,
616
+ "loss": 1.0594438314437866,
617
+ "step": 19712
618
+ },
619
+ {
620
+ "epoch": 0.9222668698905362,
621
+ "grad_norm": 1.4137084484100342,
622
+ "learning_rate": 1.557711553001523e-05,
623
+ "loss": 1.0580213069915771,
624
+ "step": 19968
625
+ },
626
+ {
627
+ "epoch": 0.9340908041199021,
628
+ "grad_norm": 0.7129232287406921,
629
+ "learning_rate": 1.1217515257622269e-05,
630
+ "loss": 1.0700064897537231,
631
+ "step": 20224
632
+ },
633
+ {
634
+ "epoch": 0.945914738349268,
635
+ "grad_norm": 2.3371353149414062,
636
+ "learning_rate": 7.565329630950746e-06,
637
+ "loss": 1.0619152784347534,
638
+ "step": 20480
639
+ },
640
+ {
641
+ "epoch": 0.945914738349268,
642
+ "eval_cos_loss": 0.2561642474507632,
643
+ "eval_loss": 1.0587050276532022,
644
+ "eval_mse_loss": 1.033088599574076,
645
+ "step": 20480
646
+ },
647
+ {
648
+ "epoch": 0.945914738349268,
649
+ "eval_cos_loss": 0.2561642474507632,
650
+ "eval_loss": 1.0587050276532022,
651
+ "eval_mse_loss": 1.033088599574076,
652
+ "eval_runtime": 26.5478,
653
+ "eval_samples_per_second": 1054.438,
654
+ "eval_steps_per_second": 16.499,
655
+ "step": 20480
656
+ }
657
+ ],
658
+ "logging_steps": 256,
659
+ "max_steps": 21651,
660
+ "num_input_tokens_seen": 0,
661
+ "num_train_epochs": 1,
662
+ "save_steps": 4096,
663
+ "stateful_callbacks": {
664
+ "TrainerControl": {
665
+ "args": {
666
+ "should_epoch_stop": false,
667
+ "should_evaluate": false,
668
+ "should_log": false,
669
+ "should_save": true,
670
+ "should_training_stop": false
671
+ },
672
+ "attributes": {}
673
+ }
674
+ },
675
+ "total_flos": 0.0,
676
+ "train_batch_size": 64,
677
+ "trial_name": null,
678
+ "trial_params": null
679
+ }
checkpoints-v5.0-discrete/checkpoint-20480/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9ead218b695fa4b63d3e4ce43bbad945367d95b659c14f235f9c2095bf1b3c31
3
+ size 5137