Hugging Face's logo

Obscure-Entropy
/
TextFlow

Safetensors
Model card Files
xet
 Community
ChiefTheLord commited on
Commit
3b1a097
·
verified ·
1 Parent(s): 08da025

Upload folder using huggingface_hub

Browse files
Files changed (9) hide show
  1. .gitattributes +1 -0
  2. checkpoints-d3.2/checkpoint-12288/eval_state.json +3 -0
  3. checkpoints-d3.2/checkpoint-12288/model.safetensors +3 -0
  4. checkpoints-d3.2/checkpoint-12288/optimizer.pt +3 -0
  5. checkpoints-d3.2/checkpoint-12288/rng_state.pth +3 -0
  6. checkpoints-d3.2/checkpoint-12288/scaler.pt +3 -0
  7. checkpoints-d3.2/checkpoint-12288/scheduler.pt +3 -0
  8. checkpoints-d3.2/checkpoint-12288/trainer_state.json +574 -0
  9. checkpoints-d3.2/checkpoint-12288/training_args.bin +3 -0
.gitattributes CHANGED
@@ -85,3 +85,4 @@ checkpoints-d1.4/checkpoint-20480/eval_state.json filter=lfs diff=lfs merge=lfs
85
  checkpoints-d3.0/checkpoint-21504/eval_state.json filter=lfs diff=lfs merge=lfs -text
86
  checkpoints-d3.1/checkpoint-8192/eval_state.json filter=lfs diff=lfs merge=lfs -text
87
  checkpoints-d3.1/checkpoint-16384/eval_state.json filter=lfs diff=lfs merge=lfs -text
 
 
85
  checkpoints-d3.0/checkpoint-21504/eval_state.json filter=lfs diff=lfs merge=lfs -text
86
  checkpoints-d3.1/checkpoint-8192/eval_state.json filter=lfs diff=lfs merge=lfs -text
87
  checkpoints-d3.1/checkpoint-16384/eval_state.json filter=lfs diff=lfs merge=lfs -text
88
+ checkpoints-d3.2/checkpoint-12288/eval_state.json filter=lfs diff=lfs merge=lfs -text
checkpoints-d3.2/checkpoint-12288/eval_state.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a9df7dd41d66cc95f551a0b122eb28d356b752e0e050dfbca8157fceaa6ea0d1
3
+ size 49099873
checkpoints-d3.2/checkpoint-12288/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:377af5eabd264a686c586017965d56a6d6c951a5980bfb94df40e27fa74e8869
3
+ size 32318104
checkpoints-d3.2/checkpoint-12288/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66d97c40c3c8fec3f4fd7d9ec69a216249765bd485c37c628ff8ff1dc4028450
3
+ size 10010635
checkpoints-d3.2/checkpoint-12288/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0c2e71208ff94e888660e624c5e753564dc55af04180be49e0d8ff6756220e37
3
+ size 14645
checkpoints-d3.2/checkpoint-12288/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ffa55f380ad889d60cf02c4b3fd1de071b4f6c3761db430565fff2faa03051d
3
+ size 1383
checkpoints-d3.2/checkpoint-12288/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:362977a3fe3e0d6b4d1f69118ac28d15e3748dcb744d952cd08de6c9c27544b6
3
+ size 1465
checkpoints-d3.2/checkpoint-12288/trainer_state.json ADDED
@@ -0,0 +1,574 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.5675488430095608,
6
+ "eval_steps": 1024,
7
+ "global_step": 12288,
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": 0.0007859937031753361,
15
+ "learning_rate": 0.000498046875,
16
+ "loss": 0.6385964751243591,
17
+ "step": 256
18
+ },
19
+ {
20
+ "epoch": 0.023647868458731697,
21
+ "grad_norm": 0.004059702157974243,
22
+ "learning_rate": 0.000998046875,
23
+ "loss": 0.6255754828453064,
24
+ "step": 512
25
+ },
26
+ {
27
+ "epoch": 0.03547180268809755,
28
+ "grad_norm": 0.0074360910803079605,
29
+ "learning_rate": 0.000999640996023194,
30
+ "loss": 0.5881161093711853,
31
+ "step": 768
32
+ },
33
+ {
34
+ "epoch": 0.047295736917463395,
35
+ "grad_norm": 0.014354221522808075,
36
+ "learning_rate": 0.0009985588674043958,
37
+ "loss": 0.5408114194869995,
38
+ "step": 1024
39
+ },
40
+ {
41
+ "epoch": 0.047295736917463395,
42
+ "eval_bleu": 0.01917810496562663,
43
+ "eval_ce_loss": 0.49575697606829205,
44
+ "eval_loss": 0.49575697606829205,
45
+ "step": 1024
46
+ },
47
+ {
48
+ "epoch": 0.047295736917463395,
49
+ "eval_bleu": 0.01917810496562663,
50
+ "eval_ce_loss": 0.49575697606829205,
51
+ "eval_loss": 0.49575697606829205,
52
+ "eval_runtime": 123.2741,
53
+ "eval_samples_per_second": 227.079,
54
+ "eval_steps_per_second": 3.553,
55
+ "step": 1024
56
+ },
57
+ {
58
+ "epoch": 0.05911967114682925,
59
+ "grad_norm": 0.01720331236720085,
60
+ "learning_rate": 0.0009967551747861387,
61
+ "loss": 0.41951486468315125,
62
+ "step": 1280
63
+ },
64
+ {
65
+ "epoch": 0.0709436053761951,
66
+ "grad_norm": 0.019904080778360367,
67
+ "learning_rate": 0.000994232528651847,
68
+ "loss": 0.32046523690223694,
69
+ "step": 1536
70
+ },
71
+ {
72
+ "epoch": 0.08276753960556095,
73
+ "grad_norm": 0.023004669696092606,
74
+ "learning_rate": 0.0009909945800260092,
75
+ "loss": 0.2763057351112366,
76
+ "step": 1792
77
+ },
78
+ {
79
+ "epoch": 0.09459147383492679,
80
+ "grad_norm": 0.025376707315444946,
81
+ "learning_rate": 0.0009870460151900522,
82
+ "loss": 0.25197452306747437,
83
+ "step": 2048
84
+ },
85
+ {
86
+ "epoch": 0.09459147383492679,
87
+ "eval_bleu": 0.15934969580648434,
88
+ "eval_ce_loss": 0.24096488823357237,
89
+ "eval_loss": 0.24096488823357237,
90
+ "step": 2048
91
+ },
92
+ {
93
+ "epoch": 0.09459147383492679,
94
+ "eval_bleu": 0.15934969580648434,
95
+ "eval_ce_loss": 0.24096488823357237,
96
+ "eval_loss": 0.24096488823357237,
97
+ "eval_runtime": 123.5565,
98
+ "eval_samples_per_second": 226.56,
99
+ "eval_steps_per_second": 3.545,
100
+ "step": 2048
101
+ },
102
+ {
103
+ "epoch": 0.10641540806429264,
104
+ "grad_norm": 0.02607455477118492,
105
+ "learning_rate": 0.0009823925488998885,
106
+ "loss": 0.23232555389404297,
107
+ "step": 2304
108
+ },
109
+ {
110
+ "epoch": 0.1182393422936585,
111
+ "grad_norm": 0.023672524839639664,
112
+ "learning_rate": 0.0009770409161149525,
113
+ "loss": 0.2173057347536087,
114
+ "step": 2560
115
+ },
116
+ {
117
+ "epoch": 0.13006327652302435,
118
+ "grad_norm": 0.03465030714869499,
119
+ "learning_rate": 0.0009709988622506973,
120
+ "loss": 0.20285777747631073,
121
+ "step": 2816
122
+ },
123
+ {
124
+ "epoch": 0.1418872107523902,
125
+ "grad_norm": 0.03454975038766861,
126
+ "learning_rate": 0.000964275131968659,
127
+ "loss": 0.19273696839809418,
128
+ "step": 3072
129
+ },
130
+ {
131
+ "epoch": 0.1418872107523902,
132
+ "eval_bleu": 0.18685399237963718,
133
+ "eval_ce_loss": 0.18767488618555678,
134
+ "eval_loss": 0.18767488618555678,
135
+ "step": 3072
136
+ },
137
+ {
138
+ "epoch": 0.1418872107523902,
139
+ "eval_bleu": 0.18685399237963718,
140
+ "eval_ce_loss": 0.18767488618555678,
141
+ "eval_loss": 0.18767488618555678,
142
+ "eval_runtime": 124.1463,
143
+ "eval_samples_per_second": 225.484,
144
+ "eval_steps_per_second": 3.528,
145
+ "step": 3072
146
+ },
147
+ {
148
+ "epoch": 0.15371114498175603,
149
+ "grad_norm": 0.028874596580863,
150
+ "learning_rate": 0.0009568794565203123,
151
+ "loss": 0.18194623291492462,
152
+ "step": 3328
153
+ },
154
+ {
155
+ "epoch": 0.1655350792111219,
156
+ "grad_norm": 0.026963606476783752,
157
+ "learning_rate": 0.0009488225396630347,
158
+ "loss": 0.1741044968366623,
159
+ "step": 3584
160
+ },
161
+ {
162
+ "epoch": 0.17735901344048774,
163
+ "grad_norm": 0.03203282505273819,
164
+ "learning_rate": 0.0009401160421685646,
165
+ "loss": 0.16683174669742584,
166
+ "step": 3840
167
+ },
168
+ {
169
+ "epoch": 0.18918294766985358,
170
+ "grad_norm": 0.02940264903008938,
171
+ "learning_rate": 0.0009307725649463714,
172
+ "loss": 0.1600237637758255,
173
+ "step": 4096
174
+ },
175
+ {
176
+ "epoch": 0.18918294766985358,
177
+ "eval_bleu": 0.2041856952873211,
178
+ "eval_ce_loss": 0.15758018299392915,
179
+ "eval_loss": 0.15758018299392915,
180
+ "step": 4096
181
+ },
182
+ {
183
+ "epoch": 0.18918294766985358,
184
+ "eval_bleu": 0.2041856952873211,
185
+ "eval_ce_loss": 0.15758018299392915,
186
+ "eval_loss": 0.15758018299392915,
187
+ "eval_runtime": 122.9443,
188
+ "eval_samples_per_second": 227.688,
189
+ "eval_steps_per_second": 3.563,
190
+ "step": 4096
191
+ },
192
+ {
193
+ "epoch": 0.20100688189921945,
194
+ "grad_norm": 0.03609512746334076,
195
+ "learning_rate": 0.0009208056308063659,
196
+ "loss": 0.15560653805732727,
197
+ "step": 4352
198
+ },
199
+ {
200
+ "epoch": 0.2128308161285853,
201
+ "grad_norm": 0.03296591714024544,
202
+ "learning_rate": 0.0009102296648873445,
203
+ "loss": 0.14899571239948273,
204
+ "step": 4608
205
+ },
206
+ {
207
+ "epoch": 0.22465475035795113,
208
+ "grad_norm": 0.034538134932518005,
209
+ "learning_rate": 0.0008990599737794927,
210
+ "loss": 0.14445939660072327,
211
+ "step": 4864
212
+ },
213
+ {
214
+ "epoch": 0.236478684587317,
215
+ "grad_norm": 0.032769329845905304,
216
+ "learning_rate": 0.0008873127233711644,
217
+ "loss": 0.13995994627475739,
218
+ "step": 5120
219
+ },
220
+ {
221
+ "epoch": 0.236478684587317,
222
+ "eval_bleu": 0.21739622472708592,
223
+ "eval_ce_loss": 0.13755980873965237,
224
+ "eval_loss": 0.13755980873965237,
225
+ "step": 5120
226
+ },
227
+ {
228
+ "epoch": 0.236478684587317,
229
+ "eval_bleu": 0.21739622472708592,
230
+ "eval_ce_loss": 0.13755980873965237,
231
+ "eval_loss": 0.13755980873965237,
232
+ "eval_runtime": 123.4264,
233
+ "eval_samples_per_second": 226.799,
234
+ "eval_steps_per_second": 3.549,
235
+ "step": 5120
236
+ },
237
+ {
238
+ "epoch": 0.24830261881668284,
239
+ "grad_norm": 0.03575948625802994,
240
+ "learning_rate": 0.0008750049154520011,
241
+ "loss": 0.13656434416770935,
242
+ "step": 5376
243
+ },
244
+ {
245
+ "epoch": 0.2601265530460487,
246
+ "grad_norm": 0.032076139003038406,
247
+ "learning_rate": 0.0008621543631062487,
248
+ "loss": 0.13251279294490814,
249
+ "step": 5632
250
+ },
251
+ {
252
+ "epoch": 0.27195048727541454,
253
+ "grad_norm": 0.03560580685734749,
254
+ "learning_rate": 0.0008487796649318904,
255
+ "loss": 0.12999674677848816,
256
+ "step": 5888
257
+ },
258
+ {
259
+ "epoch": 0.2837744215047804,
260
+ "grad_norm": 0.03557967767119408,
261
+ "learning_rate": 0.0008349001781229053,
262
+ "loss": 0.12719185650348663,
263
+ "step": 6144
264
+ },
265
+ {
266
+ "epoch": 0.2837744215047804,
267
+ "eval_bleu": 0.22349092475951987,
268
+ "eval_ce_loss": 0.1254711460032964,
269
+ "eval_loss": 0.1254711460032964,
270
+ "step": 6144
271
+ },
272
+ {
273
+ "epoch": 0.2837744215047804,
274
+ "eval_bleu": 0.22349092475951987,
275
+ "eval_ce_loss": 0.1254711460032964,
276
+ "eval_loss": 0.1254711460032964,
277
+ "eval_runtime": 124.846,
278
+ "eval_samples_per_second": 224.22,
279
+ "eval_steps_per_second": 3.508,
280
+ "step": 6144
281
+ },
282
+ {
283
+ "epoch": 0.2955983557341462,
284
+ "grad_norm": 0.03403434529900551,
285
+ "learning_rate": 0.0008205359904536107,
286
+ "loss": 0.12399476021528244,
287
+ "step": 6400
288
+ },
289
+ {
290
+ "epoch": 0.30742228996351206,
291
+ "grad_norm": 0.03340385481715202,
292
+ "learning_rate": 0.0008057078912056363,
293
+ "loss": 0.12178181111812592,
294
+ "step": 6656
295
+ },
296
+ {
297
+ "epoch": 0.3192462241928779,
298
+ "grad_norm": 0.033506471663713455,
299
+ "learning_rate": 0.0007904373410796086,
300
+ "loss": 0.11948449909687042,
301
+ "step": 6912
302
+ },
303
+ {
304
+ "epoch": 0.3310701584222438,
305
+ "grad_norm": 0.0328289233148098,
306
+ "learning_rate": 0.0007747464411350876,
307
+ "loss": 0.11748082935810089,
308
+ "step": 7168
309
+ },
310
+ {
311
+ "epoch": 0.3310701584222438,
312
+ "eval_bleu": 0.22974485130663763,
313
+ "eval_ce_loss": 0.11694975512008689,
314
+ "eval_loss": 0.11694975512008689,
315
+ "step": 7168
316
+ },
317
+ {
318
+ "epoch": 0.3310701584222438,
319
+ "eval_bleu": 0.22974485130663763,
320
+ "eval_ce_loss": 0.11694975512008689,
321
+ "eval_loss": 0.11694975512008689,
322
+ "eval_runtime": 126.6834,
323
+ "eval_samples_per_second": 220.968,
324
+ "eval_steps_per_second": 3.457,
325
+ "step": 7168
326
+ },
327
+ {
328
+ "epoch": 0.34289409265160964,
329
+ "grad_norm": 0.04062662646174431,
330
+ "learning_rate": 0.000758657900803716,
331
+ "loss": 0.11517409980297089,
332
+ "step": 7424
333
+ },
334
+ {
335
+ "epoch": 0.3547180268809755,
336
+ "grad_norm": 0.03233740106225014,
337
+ "learning_rate": 0.000742195005021869,
338
+ "loss": 0.11353136599063873,
339
+ "step": 7680
340
+ },
341
+ {
342
+ "epoch": 0.3665419611103413,
343
+ "grad_norm": 0.04508234187960625,
344
+ "learning_rate": 0.0007253815805303786,
345
+ "loss": 0.11149852722883224,
346
+ "step": 7936
347
+ },
348
+ {
349
+ "epoch": 0.37836589533970716,
350
+ "grad_norm": 0.0339302197098732,
351
+ "learning_rate": 0.0007082419613901028,
352
+ "loss": 0.10967232286930084,
353
+ "step": 8192
354
+ },
355
+ {
356
+ "epoch": 0.37836589533970716,
357
+ "eval_bleu": 0.23402933156555222,
358
+ "eval_ce_loss": 0.10964651824268576,
359
+ "eval_loss": 0.10964651824268576,
360
+ "step": 8192
361
+ },
362
+ {
363
+ "epoch": 0.37836589533970716,
364
+ "eval_bleu": 0.23402933156555222,
365
+ "eval_ce_loss": 0.10964651824268576,
366
+ "eval_loss": 0.10964651824268576,
367
+ "eval_runtime": 126.1015,
368
+ "eval_samples_per_second": 221.988,
369
+ "eval_steps_per_second": 3.473,
370
+ "step": 8192
371
+ },
372
+ {
373
+ "epoch": 0.390189829569073,
374
+ "grad_norm": 0.03507415950298309,
375
+ "learning_rate": 0.0006908009537632514,
376
+ "loss": 0.10870054364204407,
377
+ "step": 8448
378
+ },
379
+ {
380
+ "epoch": 0.4020137637984389,
381
+ "grad_norm": 0.03351368010044098,
382
+ "learning_rate": 0.0006730838000114403,
383
+ "loss": 0.10755178332328796,
384
+ "step": 8704
385
+ },
386
+ {
387
+ "epoch": 0.41383769802780473,
388
+ "grad_norm": 0.03829547390341759,
389
+ "learning_rate": 0.0006551161421624341,
390
+ "loss": 0.1055288016796112,
391
+ "step": 8960
392
+ },
393
+ {
394
+ "epoch": 0.4256616322571706,
395
+ "grad_norm": 0.029935523867607117,
396
+ "learning_rate": 0.0006369239847984517,
397
+ "loss": 0.10448558628559113,
398
+ "step": 9216
399
+ },
400
+ {
401
+ "epoch": 0.4256616322571706,
402
+ "eval_bleu": 0.23801999524762174,
403
+ "eval_ce_loss": 0.10415493887445154,
404
+ "eval_loss": 0.10415493887445154,
405
+ "step": 9216
406
+ },
407
+ {
408
+ "epoch": 0.4256616322571706,
409
+ "eval_bleu": 0.23801999524762174,
410
+ "eval_ce_loss": 0.10415493887445154,
411
+ "eval_loss": 0.10415493887445154,
412
+ "eval_runtime": 127.389,
413
+ "eval_samples_per_second": 219.744,
414
+ "eval_steps_per_second": 3.438,
415
+ "step": 9216
416
+ },
417
+ {
418
+ "epoch": 0.4374855664865364,
419
+ "grad_norm": 0.029414288699626923,
420
+ "learning_rate": 0.0006185336574197479,
421
+ "loss": 0.10328588634729385,
422
+ "step": 9472
423
+ },
424
+ {
425
+ "epoch": 0.44930950071590225,
426
+ "grad_norm": 0.03512301295995712,
427
+ "learning_rate": 0.0005999717763379407,
428
+ "loss": 0.10299594700336456,
429
+ "step": 9728
430
+ },
431
+ {
432
+ "epoch": 0.4611334349452681,
433
+ "grad_norm": 0.04324057698249817,
434
+ "learning_rate": 0.0005812652061542363,
435
+ "loss": 0.10201191902160645,
436
+ "step": 9984
437
+ },
438
+ {
439
+ "epoch": 0.472957369174634,
440
+ "grad_norm": 0.032921064645051956,
441
+ "learning_rate": 0.0005624410208783071,
442
+ "loss": 0.10033029317855835,
443
+ "step": 10240
444
+ },
445
+ {
446
+ "epoch": 0.472957369174634,
447
+ "eval_bleu": 0.23928028701230103,
448
+ "eval_ce_loss": 0.10052529907866156,
449
+ "eval_loss": 0.10052529907866156,
450
+ "step": 10240
451
+ },
452
+ {
453
+ "epoch": 0.472957369174634,
454
+ "eval_bleu": 0.23928028701230103,
455
+ "eval_ce_loss": 0.10052529907866156,
456
+ "eval_loss": 0.10052529907866156,
457
+ "eval_runtime": 126.6952,
458
+ "eval_samples_per_second": 220.948,
459
+ "eval_steps_per_second": 3.457,
460
+ "step": 10240
461
+ },
462
+ {
463
+ "epoch": 0.48478130340399983,
464
+ "grad_norm": 0.038171831518411636,
465
+ "learning_rate": 0.0005435264647440881,
466
+ "loss": 0.09996124356985092,
467
+ "step": 10496
468
+ },
469
+ {
470
+ "epoch": 0.49660523763336567,
471
+ "grad_norm": 0.03536969795823097,
472
+ "learning_rate": 0.000524548912779213,
473
+ "loss": 0.09928364306688309,
474
+ "step": 10752
475
+ },
476
+ {
477
+ "epoch": 0.5084291718627315,
478
+ "grad_norm": 0.03323720768094063,
479
+ "learning_rate": 0.0005055358311851499,
480
+ "loss": 0.09906301647424698,
481
+ "step": 11008
482
+ },
483
+ {
484
+ "epoch": 0.5202531060920974,
485
+ "grad_norm": 0.033888787031173706,
486
+ "learning_rate": 0.0004865147375853812,
487
+ "loss": 0.09757085144519806,
488
+ "step": 11264
489
+ },
490
+ {
491
+ "epoch": 0.5202531060920974,
492
+ "eval_bleu": 0.24364428948892838,
493
+ "eval_ce_loss": 0.09666734056113517,
494
+ "eval_loss": 0.09666734056113517,
495
+ "step": 11264
496
+ },
497
+ {
498
+ "epoch": 0.5202531060920974,
499
+ "eval_bleu": 0.24364428948892838,
500
+ "eval_ce_loss": 0.09666734056113517,
501
+ "eval_loss": 0.09666734056113517,
502
+ "eval_runtime": 124.874,
503
+ "eval_samples_per_second": 224.17,
504
+ "eval_steps_per_second": 3.508,
505
+ "step": 11264
506
+ },
507
+ {
508
+ "epoch": 0.5320770403214632,
509
+ "grad_norm": 0.03228915110230446,
510
+ "learning_rate": 0.0004675131611991607,
511
+ "loss": 0.09719787538051605,
512
+ "step": 11520
513
+ },
514
+ {
515
+ "epoch": 0.5439009745508291,
516
+ "grad_norm": 0.03569363057613373,
517
+ "learning_rate": 0.0004485586029984899,
518
+ "loss": 0.09681548178195953,
519
+ "step": 11776
520
+ },
521
+ {
522
+ "epoch": 0.5557249087801949,
523
+ "grad_norm": 0.03645534813404083,
524
+ "learning_rate": 0.00042967849590597266,
525
+ "loss": 0.09616198390722275,
526
+ "step": 12032
527
+ },
528
+ {
529
+ "epoch": 0.5675488430095608,
530
+ "grad_norm": 0.03306020796298981,
531
+ "learning_rate": 0.0004109001650911621,
532
+ "loss": 0.09529497474431992,
533
+ "step": 12288
534
+ },
535
+ {
536
+ "epoch": 0.5675488430095608,
537
+ "eval_bleu": 0.24421693548361179,
538
+ "eval_ce_loss": 0.09502403213553233,
539
+ "eval_loss": 0.09502403213553233,
540
+ "step": 12288
541
+ },
542
+ {
543
+ "epoch": 0.5675488430095608,
544
+ "eval_bleu": 0.24421693548361179,
545
+ "eval_ce_loss": 0.09502403213553233,
546
+ "eval_loss": 0.09502403213553233,
547
+ "eval_runtime": 126.0135,
548
+ "eval_samples_per_second": 222.143,
549
+ "eval_steps_per_second": 3.476,
550
+ "step": 12288
551
+ }
552
+ ],
553
+ "logging_steps": 256,
554
+ "max_steps": 21651,
555
+ "num_input_tokens_seen": 0,
556
+ "num_train_epochs": 1,
557
+ "save_steps": 1024,
558
+ "stateful_callbacks": {
559
+ "TrainerControl": {
560
+ "args": {
561
+ "should_epoch_stop": false,
562
+ "should_evaluate": false,
563
+ "should_log": false,
564
+ "should_save": true,
565
+ "should_training_stop": false
566
+ },
567
+ "attributes": {}
568
+ }
569
+ },
570
+ "total_flos": 0.0,
571
+ "train_batch_size": 64,
572
+ "trial_name": null,
573
+ "trial_params": null
574
+ }
checkpoints-d3.2/checkpoint-12288/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:61a16bb839f687414b8e48611327c4b9cfddeefe38c031ca70808f9a97c476b7
3
+ size 5137