Rafeq commited on
Commit
8ed3e5d
·
1 Parent(s): 7ee99cc

Upload 6 files

Browse files
Files changed (5) hide show
  1. rng_state.pth +3 -0
  2. scaler.pt +3 -0
  3. scheduler.pt +3 -0
  4. trainer_state.json +1567 -0
  5. training_args.bin +3 -0
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d15474ca90b892e86d1066962d0f482c1edde69b8b9995670e2d92478c97d860
3
+ size 14639
scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd455d951f5e613dfc9a1bd1da593f0b36fa8adf4ab0abda4f1be3af9fdeb1bf
3
+ size 557
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:02cca9800653b01780ed3162cf9f858958cbef4b32bc892af48c15ff853fdb69
3
+ size 627
trainer_state.json ADDED
@@ -0,0 +1,1567 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 4.9890764647467725,
5
+ "global_step": 2510,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.02,
12
+ "learning_rate": 9.960238568588469e-05,
13
+ "loss": 0.1445,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.04,
18
+ "learning_rate": 9.924453280318093e-05,
19
+ "loss": 0.3715,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.06,
24
+ "learning_rate": 9.884691848906562e-05,
25
+ "loss": 0.5201,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.08,
30
+ "learning_rate": 9.84493041749503e-05,
31
+ "loss": 0.355,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.1,
36
+ "learning_rate": 9.805168986083499e-05,
37
+ "loss": 0.4739,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.12,
42
+ "learning_rate": 9.765407554671967e-05,
43
+ "loss": 0.3795,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.14,
48
+ "learning_rate": 9.725646123260437e-05,
49
+ "loss": 0.2449,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.16,
54
+ "learning_rate": 9.685884691848906e-05,
55
+ "loss": 0.2834,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.18,
60
+ "learning_rate": 9.646123260437376e-05,
61
+ "loss": 0.4803,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.2,
66
+ "learning_rate": 9.606361829025845e-05,
67
+ "loss": 0.4239,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.22,
72
+ "learning_rate": 9.566600397614315e-05,
73
+ "loss": 0.4822,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.24,
78
+ "learning_rate": 9.530815109343937e-05,
79
+ "loss": 0.6423,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.26,
84
+ "learning_rate": 9.491053677932406e-05,
85
+ "loss": 0.4146,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.28,
90
+ "learning_rate": 9.451292246520876e-05,
91
+ "loss": 0.3988,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.3,
96
+ "learning_rate": 9.411530815109344e-05,
97
+ "loss": 0.576,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.32,
102
+ "learning_rate": 9.371769383697814e-05,
103
+ "loss": 0.3698,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.34,
108
+ "learning_rate": 9.332007952286283e-05,
109
+ "loss": 0.4959,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.36,
114
+ "learning_rate": 9.292246520874752e-05,
115
+ "loss": 0.353,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.38,
120
+ "learning_rate": 9.25248508946322e-05,
121
+ "loss": 0.3502,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.4,
126
+ "learning_rate": 9.21272365805169e-05,
127
+ "loss": 0.2586,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.42,
132
+ "learning_rate": 9.172962226640159e-05,
133
+ "loss": 0.3487,
134
+ "step": 210
135
+ },
136
+ {
137
+ "epoch": 0.44,
138
+ "learning_rate": 9.133200795228629e-05,
139
+ "loss": 0.2338,
140
+ "step": 220
141
+ },
142
+ {
143
+ "epoch": 0.46,
144
+ "learning_rate": 9.093439363817098e-05,
145
+ "loss": 0.3604,
146
+ "step": 230
147
+ },
148
+ {
149
+ "epoch": 0.48,
150
+ "learning_rate": 9.053677932405566e-05,
151
+ "loss": 0.1628,
152
+ "step": 240
153
+ },
154
+ {
155
+ "epoch": 0.5,
156
+ "learning_rate": 9.013916500994036e-05,
157
+ "loss": 0.3106,
158
+ "step": 250
159
+ },
160
+ {
161
+ "epoch": 0.52,
162
+ "learning_rate": 8.974155069582505e-05,
163
+ "loss": 0.3068,
164
+ "step": 260
165
+ },
166
+ {
167
+ "epoch": 0.54,
168
+ "learning_rate": 8.934393638170975e-05,
169
+ "loss": 0.273,
170
+ "step": 270
171
+ },
172
+ {
173
+ "epoch": 0.56,
174
+ "learning_rate": 8.894632206759443e-05,
175
+ "loss": 0.1321,
176
+ "step": 280
177
+ },
178
+ {
179
+ "epoch": 0.58,
180
+ "learning_rate": 8.854870775347913e-05,
181
+ "loss": 0.4806,
182
+ "step": 290
183
+ },
184
+ {
185
+ "epoch": 0.6,
186
+ "learning_rate": 8.815109343936382e-05,
187
+ "loss": 0.1793,
188
+ "step": 300
189
+ },
190
+ {
191
+ "epoch": 0.62,
192
+ "learning_rate": 8.775347912524852e-05,
193
+ "loss": 0.2027,
194
+ "step": 310
195
+ },
196
+ {
197
+ "epoch": 0.64,
198
+ "learning_rate": 8.735586481113321e-05,
199
+ "loss": 0.3706,
200
+ "step": 320
201
+ },
202
+ {
203
+ "epoch": 0.66,
204
+ "learning_rate": 8.695825049701791e-05,
205
+ "loss": 0.4457,
206
+ "step": 330
207
+ },
208
+ {
209
+ "epoch": 0.68,
210
+ "learning_rate": 8.65606361829026e-05,
211
+ "loss": 0.3953,
212
+ "step": 340
213
+ },
214
+ {
215
+ "epoch": 0.7,
216
+ "learning_rate": 8.616302186878728e-05,
217
+ "loss": 0.132,
218
+ "step": 350
219
+ },
220
+ {
221
+ "epoch": 0.71,
222
+ "learning_rate": 8.576540755467197e-05,
223
+ "loss": 0.3439,
224
+ "step": 360
225
+ },
226
+ {
227
+ "epoch": 0.73,
228
+ "learning_rate": 8.536779324055667e-05,
229
+ "loss": 0.6228,
230
+ "step": 370
231
+ },
232
+ {
233
+ "epoch": 0.75,
234
+ "learning_rate": 8.497017892644135e-05,
235
+ "loss": 0.3251,
236
+ "step": 380
237
+ },
238
+ {
239
+ "epoch": 0.77,
240
+ "learning_rate": 8.457256461232604e-05,
241
+ "loss": 0.3342,
242
+ "step": 390
243
+ },
244
+ {
245
+ "epoch": 0.79,
246
+ "learning_rate": 8.417495029821074e-05,
247
+ "loss": 0.316,
248
+ "step": 400
249
+ },
250
+ {
251
+ "epoch": 0.81,
252
+ "learning_rate": 8.377733598409543e-05,
253
+ "loss": 0.2536,
254
+ "step": 410
255
+ },
256
+ {
257
+ "epoch": 0.83,
258
+ "learning_rate": 8.337972166998013e-05,
259
+ "loss": 0.2428,
260
+ "step": 420
261
+ },
262
+ {
263
+ "epoch": 0.85,
264
+ "learning_rate": 8.298210735586481e-05,
265
+ "loss": 0.1508,
266
+ "step": 430
267
+ },
268
+ {
269
+ "epoch": 0.87,
270
+ "learning_rate": 8.258449304174951e-05,
271
+ "loss": 0.2234,
272
+ "step": 440
273
+ },
274
+ {
275
+ "epoch": 0.89,
276
+ "learning_rate": 8.21868787276342e-05,
277
+ "loss": 0.3659,
278
+ "step": 450
279
+ },
280
+ {
281
+ "epoch": 0.91,
282
+ "learning_rate": 8.17892644135189e-05,
283
+ "loss": 0.2216,
284
+ "step": 460
285
+ },
286
+ {
287
+ "epoch": 0.93,
288
+ "learning_rate": 8.139165009940358e-05,
289
+ "loss": 0.3196,
290
+ "step": 470
291
+ },
292
+ {
293
+ "epoch": 0.95,
294
+ "learning_rate": 8.099403578528827e-05,
295
+ "loss": 0.3921,
296
+ "step": 480
297
+ },
298
+ {
299
+ "epoch": 0.97,
300
+ "learning_rate": 8.059642147117297e-05,
301
+ "loss": 0.3935,
302
+ "step": 490
303
+ },
304
+ {
305
+ "epoch": 0.99,
306
+ "learning_rate": 8.019880715705766e-05,
307
+ "loss": 0.1749,
308
+ "step": 500
309
+ },
310
+ {
311
+ "epoch": 0.99,
312
+ "eval_accuracy": 0.9107142686843872,
313
+ "eval_loss": 0.2552814781665802,
314
+ "eval_runtime": 19.16,
315
+ "eval_samples_per_second": 26.305,
316
+ "eval_steps_per_second": 6.576,
317
+ "step": 500
318
+ },
319
+ {
320
+ "epoch": 1.01,
321
+ "learning_rate": 7.980119284294236e-05,
322
+ "loss": 0.3414,
323
+ "step": 510
324
+ },
325
+ {
326
+ "epoch": 1.03,
327
+ "learning_rate": 7.940357852882704e-05,
328
+ "loss": 0.2059,
329
+ "step": 520
330
+ },
331
+ {
332
+ "epoch": 1.05,
333
+ "learning_rate": 7.900596421471174e-05,
334
+ "loss": 0.1762,
335
+ "step": 530
336
+ },
337
+ {
338
+ "epoch": 1.07,
339
+ "learning_rate": 7.860834990059643e-05,
340
+ "loss": 0.118,
341
+ "step": 540
342
+ },
343
+ {
344
+ "epoch": 1.09,
345
+ "learning_rate": 7.821073558648112e-05,
346
+ "loss": 0.1524,
347
+ "step": 550
348
+ },
349
+ {
350
+ "epoch": 1.11,
351
+ "learning_rate": 7.78131212723658e-05,
352
+ "loss": 0.1576,
353
+ "step": 560
354
+ },
355
+ {
356
+ "epoch": 1.13,
357
+ "learning_rate": 7.74155069582505e-05,
358
+ "loss": 0.2975,
359
+ "step": 570
360
+ },
361
+ {
362
+ "epoch": 1.15,
363
+ "learning_rate": 7.701789264413519e-05,
364
+ "loss": 0.2985,
365
+ "step": 580
366
+ },
367
+ {
368
+ "epoch": 1.17,
369
+ "learning_rate": 7.662027833001987e-05,
370
+ "loss": 0.1437,
371
+ "step": 590
372
+ },
373
+ {
374
+ "epoch": 1.19,
375
+ "learning_rate": 7.622266401590457e-05,
376
+ "loss": 0.2702,
377
+ "step": 600
378
+ },
379
+ {
380
+ "epoch": 1.21,
381
+ "learning_rate": 7.582504970178926e-05,
382
+ "loss": 0.2913,
383
+ "step": 610
384
+ },
385
+ {
386
+ "epoch": 1.23,
387
+ "learning_rate": 7.542743538767396e-05,
388
+ "loss": 0.1741,
389
+ "step": 620
390
+ },
391
+ {
392
+ "epoch": 1.25,
393
+ "learning_rate": 7.502982107355865e-05,
394
+ "loss": 0.1987,
395
+ "step": 630
396
+ },
397
+ {
398
+ "epoch": 1.27,
399
+ "learning_rate": 7.463220675944335e-05,
400
+ "loss": 0.2568,
401
+ "step": 640
402
+ },
403
+ {
404
+ "epoch": 1.29,
405
+ "learning_rate": 7.423459244532803e-05,
406
+ "loss": 0.4236,
407
+ "step": 650
408
+ },
409
+ {
410
+ "epoch": 1.31,
411
+ "learning_rate": 7.383697813121273e-05,
412
+ "loss": 0.1788,
413
+ "step": 660
414
+ },
415
+ {
416
+ "epoch": 1.33,
417
+ "learning_rate": 7.343936381709742e-05,
418
+ "loss": 0.2973,
419
+ "step": 670
420
+ },
421
+ {
422
+ "epoch": 1.35,
423
+ "learning_rate": 7.304174950298212e-05,
424
+ "loss": 0.3244,
425
+ "step": 680
426
+ },
427
+ {
428
+ "epoch": 1.37,
429
+ "learning_rate": 7.26441351888668e-05,
430
+ "loss": 0.2661,
431
+ "step": 690
432
+ },
433
+ {
434
+ "epoch": 1.39,
435
+ "learning_rate": 7.22465208747515e-05,
436
+ "loss": 0.3832,
437
+ "step": 700
438
+ },
439
+ {
440
+ "epoch": 1.41,
441
+ "learning_rate": 7.184890656063619e-05,
442
+ "loss": 0.1453,
443
+ "step": 710
444
+ },
445
+ {
446
+ "epoch": 1.43,
447
+ "learning_rate": 7.145129224652088e-05,
448
+ "loss": 0.1034,
449
+ "step": 720
450
+ },
451
+ {
452
+ "epoch": 1.45,
453
+ "learning_rate": 7.105367793240556e-05,
454
+ "loss": 0.2906,
455
+ "step": 730
456
+ },
457
+ {
458
+ "epoch": 1.47,
459
+ "learning_rate": 7.065606361829026e-05,
460
+ "loss": 0.2115,
461
+ "step": 740
462
+ },
463
+ {
464
+ "epoch": 1.49,
465
+ "learning_rate": 7.025844930417495e-05,
466
+ "loss": 0.3433,
467
+ "step": 750
468
+ },
469
+ {
470
+ "epoch": 1.51,
471
+ "learning_rate": 6.986083499005964e-05,
472
+ "loss": 0.1599,
473
+ "step": 760
474
+ },
475
+ {
476
+ "epoch": 1.53,
477
+ "learning_rate": 6.946322067594434e-05,
478
+ "loss": 0.1725,
479
+ "step": 770
480
+ },
481
+ {
482
+ "epoch": 1.55,
483
+ "learning_rate": 6.910536779324056e-05,
484
+ "loss": 0.428,
485
+ "step": 780
486
+ },
487
+ {
488
+ "epoch": 1.57,
489
+ "learning_rate": 6.870775347912525e-05,
490
+ "loss": 0.1033,
491
+ "step": 790
492
+ },
493
+ {
494
+ "epoch": 1.59,
495
+ "learning_rate": 6.831013916500995e-05,
496
+ "loss": 0.314,
497
+ "step": 800
498
+ },
499
+ {
500
+ "epoch": 1.61,
501
+ "learning_rate": 6.791252485089463e-05,
502
+ "loss": 0.146,
503
+ "step": 810
504
+ },
505
+ {
506
+ "epoch": 1.63,
507
+ "learning_rate": 6.751491053677933e-05,
508
+ "loss": 0.181,
509
+ "step": 820
510
+ },
511
+ {
512
+ "epoch": 1.65,
513
+ "learning_rate": 6.711729622266402e-05,
514
+ "loss": 0.2174,
515
+ "step": 830
516
+ },
517
+ {
518
+ "epoch": 1.67,
519
+ "learning_rate": 6.671968190854872e-05,
520
+ "loss": 0.1458,
521
+ "step": 840
522
+ },
523
+ {
524
+ "epoch": 1.69,
525
+ "learning_rate": 6.632206759443341e-05,
526
+ "loss": 0.1477,
527
+ "step": 850
528
+ },
529
+ {
530
+ "epoch": 1.71,
531
+ "learning_rate": 6.59244532803181e-05,
532
+ "loss": 0.3046,
533
+ "step": 860
534
+ },
535
+ {
536
+ "epoch": 1.73,
537
+ "learning_rate": 6.552683896620278e-05,
538
+ "loss": 0.2796,
539
+ "step": 870
540
+ },
541
+ {
542
+ "epoch": 1.75,
543
+ "learning_rate": 6.512922465208748e-05,
544
+ "loss": 0.1284,
545
+ "step": 880
546
+ },
547
+ {
548
+ "epoch": 1.77,
549
+ "learning_rate": 6.473161033797217e-05,
550
+ "loss": 0.3014,
551
+ "step": 890
552
+ },
553
+ {
554
+ "epoch": 1.79,
555
+ "learning_rate": 6.433399602385685e-05,
556
+ "loss": 0.2031,
557
+ "step": 900
558
+ },
559
+ {
560
+ "epoch": 1.81,
561
+ "learning_rate": 6.393638170974155e-05,
562
+ "loss": 0.134,
563
+ "step": 910
564
+ },
565
+ {
566
+ "epoch": 1.83,
567
+ "learning_rate": 6.353876739562624e-05,
568
+ "loss": 0.2085,
569
+ "step": 920
570
+ },
571
+ {
572
+ "epoch": 1.85,
573
+ "learning_rate": 6.314115308151094e-05,
574
+ "loss": 0.2496,
575
+ "step": 930
576
+ },
577
+ {
578
+ "epoch": 1.87,
579
+ "learning_rate": 6.274353876739562e-05,
580
+ "loss": 0.1984,
581
+ "step": 940
582
+ },
583
+ {
584
+ "epoch": 1.89,
585
+ "learning_rate": 6.234592445328032e-05,
586
+ "loss": 0.2887,
587
+ "step": 950
588
+ },
589
+ {
590
+ "epoch": 1.91,
591
+ "learning_rate": 6.194831013916501e-05,
592
+ "loss": 0.1923,
593
+ "step": 960
594
+ },
595
+ {
596
+ "epoch": 1.93,
597
+ "learning_rate": 6.155069582504971e-05,
598
+ "loss": 0.1373,
599
+ "step": 970
600
+ },
601
+ {
602
+ "epoch": 1.95,
603
+ "learning_rate": 6.11530815109344e-05,
604
+ "loss": 0.2089,
605
+ "step": 980
606
+ },
607
+ {
608
+ "epoch": 1.97,
609
+ "learning_rate": 6.075546719681909e-05,
610
+ "loss": 0.2066,
611
+ "step": 990
612
+ },
613
+ {
614
+ "epoch": 1.99,
615
+ "learning_rate": 6.035785288270378e-05,
616
+ "loss": 0.1106,
617
+ "step": 1000
618
+ },
619
+ {
620
+ "epoch": 1.99,
621
+ "eval_accuracy": 0.932539701461792,
622
+ "eval_loss": 0.25381073355674744,
623
+ "eval_runtime": 20.2105,
624
+ "eval_samples_per_second": 24.938,
625
+ "eval_steps_per_second": 6.234,
626
+ "step": 1000
627
+ },
628
+ {
629
+ "epoch": 2.01,
630
+ "learning_rate": 5.996023856858847e-05,
631
+ "loss": 0.125,
632
+ "step": 1010
633
+ },
634
+ {
635
+ "epoch": 2.03,
636
+ "learning_rate": 5.956262425447316e-05,
637
+ "loss": 0.1804,
638
+ "step": 1020
639
+ },
640
+ {
641
+ "epoch": 2.05,
642
+ "learning_rate": 5.916500994035785e-05,
643
+ "loss": 0.0419,
644
+ "step": 1030
645
+ },
646
+ {
647
+ "epoch": 2.07,
648
+ "learning_rate": 5.876739562624255e-05,
649
+ "loss": 0.1823,
650
+ "step": 1040
651
+ },
652
+ {
653
+ "epoch": 2.09,
654
+ "learning_rate": 5.8369781312127236e-05,
655
+ "loss": 0.085,
656
+ "step": 1050
657
+ },
658
+ {
659
+ "epoch": 2.11,
660
+ "learning_rate": 5.7972166998011936e-05,
661
+ "loss": 0.5209,
662
+ "step": 1060
663
+ },
664
+ {
665
+ "epoch": 2.13,
666
+ "learning_rate": 5.757455268389662e-05,
667
+ "loss": 0.1744,
668
+ "step": 1070
669
+ },
670
+ {
671
+ "epoch": 2.15,
672
+ "learning_rate": 5.717693836978132e-05,
673
+ "loss": 0.1288,
674
+ "step": 1080
675
+ },
676
+ {
677
+ "epoch": 2.17,
678
+ "learning_rate": 5.677932405566601e-05,
679
+ "loss": 0.0729,
680
+ "step": 1090
681
+ },
682
+ {
683
+ "epoch": 2.19,
684
+ "learning_rate": 5.63817097415507e-05,
685
+ "loss": 0.1138,
686
+ "step": 1100
687
+ },
688
+ {
689
+ "epoch": 2.21,
690
+ "learning_rate": 5.598409542743539e-05,
691
+ "loss": 0.2507,
692
+ "step": 1110
693
+ },
694
+ {
695
+ "epoch": 2.23,
696
+ "learning_rate": 5.5586481113320074e-05,
697
+ "loss": 0.2038,
698
+ "step": 1120
699
+ },
700
+ {
701
+ "epoch": 2.25,
702
+ "learning_rate": 5.5188866799204774e-05,
703
+ "loss": 0.3174,
704
+ "step": 1130
705
+ },
706
+ {
707
+ "epoch": 2.27,
708
+ "learning_rate": 5.479125248508946e-05,
709
+ "loss": 0.2045,
710
+ "step": 1140
711
+ },
712
+ {
713
+ "epoch": 2.29,
714
+ "learning_rate": 5.439363817097416e-05,
715
+ "loss": 0.1042,
716
+ "step": 1150
717
+ },
718
+ {
719
+ "epoch": 2.31,
720
+ "learning_rate": 5.399602385685885e-05,
721
+ "loss": 0.0531,
722
+ "step": 1160
723
+ },
724
+ {
725
+ "epoch": 2.33,
726
+ "learning_rate": 5.359840954274355e-05,
727
+ "loss": 0.1353,
728
+ "step": 1170
729
+ },
730
+ {
731
+ "epoch": 2.35,
732
+ "learning_rate": 5.320079522862823e-05,
733
+ "loss": 0.3447,
734
+ "step": 1180
735
+ },
736
+ {
737
+ "epoch": 2.37,
738
+ "learning_rate": 5.2803180914512926e-05,
739
+ "loss": 0.1704,
740
+ "step": 1190
741
+ },
742
+ {
743
+ "epoch": 2.39,
744
+ "learning_rate": 5.240556660039761e-05,
745
+ "loss": 0.2148,
746
+ "step": 1200
747
+ },
748
+ {
749
+ "epoch": 2.41,
750
+ "learning_rate": 5.200795228628231e-05,
751
+ "loss": 0.222,
752
+ "step": 1210
753
+ },
754
+ {
755
+ "epoch": 2.43,
756
+ "learning_rate": 5.1610337972167e-05,
757
+ "loss": 0.0645,
758
+ "step": 1220
759
+ },
760
+ {
761
+ "epoch": 2.44,
762
+ "learning_rate": 5.12127236580517e-05,
763
+ "loss": 0.1284,
764
+ "step": 1230
765
+ },
766
+ {
767
+ "epoch": 2.46,
768
+ "learning_rate": 5.0815109343936385e-05,
769
+ "loss": 0.1021,
770
+ "step": 1240
771
+ },
772
+ {
773
+ "epoch": 2.48,
774
+ "learning_rate": 5.041749502982107e-05,
775
+ "loss": 0.1881,
776
+ "step": 1250
777
+ },
778
+ {
779
+ "epoch": 2.5,
780
+ "learning_rate": 5.001988071570577e-05,
781
+ "loss": 0.1158,
782
+ "step": 1260
783
+ },
784
+ {
785
+ "epoch": 2.52,
786
+ "learning_rate": 4.9622266401590465e-05,
787
+ "loss": 0.2495,
788
+ "step": 1270
789
+ },
790
+ {
791
+ "epoch": 2.54,
792
+ "learning_rate": 4.922465208747515e-05,
793
+ "loss": 0.0767,
794
+ "step": 1280
795
+ },
796
+ {
797
+ "epoch": 2.56,
798
+ "learning_rate": 4.882703777335984e-05,
799
+ "loss": 0.1401,
800
+ "step": 1290
801
+ },
802
+ {
803
+ "epoch": 2.58,
804
+ "learning_rate": 4.842942345924453e-05,
805
+ "loss": 0.1357,
806
+ "step": 1300
807
+ },
808
+ {
809
+ "epoch": 2.6,
810
+ "learning_rate": 4.8031809145129224e-05,
811
+ "loss": 0.1962,
812
+ "step": 1310
813
+ },
814
+ {
815
+ "epoch": 2.62,
816
+ "learning_rate": 4.763419483101392e-05,
817
+ "loss": 0.1744,
818
+ "step": 1320
819
+ },
820
+ {
821
+ "epoch": 2.64,
822
+ "learning_rate": 4.723658051689861e-05,
823
+ "loss": 0.1815,
824
+ "step": 1330
825
+ },
826
+ {
827
+ "epoch": 2.66,
828
+ "learning_rate": 4.68389662027833e-05,
829
+ "loss": 0.275,
830
+ "step": 1340
831
+ },
832
+ {
833
+ "epoch": 2.68,
834
+ "learning_rate": 4.6441351888667996e-05,
835
+ "loss": 0.1725,
836
+ "step": 1350
837
+ },
838
+ {
839
+ "epoch": 2.7,
840
+ "learning_rate": 4.604373757455269e-05,
841
+ "loss": 0.1071,
842
+ "step": 1360
843
+ },
844
+ {
845
+ "epoch": 2.72,
846
+ "learning_rate": 4.5646123260437376e-05,
847
+ "loss": 0.0908,
848
+ "step": 1370
849
+ },
850
+ {
851
+ "epoch": 2.74,
852
+ "learning_rate": 4.524850894632207e-05,
853
+ "loss": 0.0793,
854
+ "step": 1380
855
+ },
856
+ {
857
+ "epoch": 2.76,
858
+ "learning_rate": 4.485089463220676e-05,
859
+ "loss": 0.114,
860
+ "step": 1390
861
+ },
862
+ {
863
+ "epoch": 2.78,
864
+ "learning_rate": 4.4453280318091455e-05,
865
+ "loss": 0.2036,
866
+ "step": 1400
867
+ },
868
+ {
869
+ "epoch": 2.8,
870
+ "learning_rate": 4.405566600397614e-05,
871
+ "loss": 0.0941,
872
+ "step": 1410
873
+ },
874
+ {
875
+ "epoch": 2.82,
876
+ "learning_rate": 4.3658051689860835e-05,
877
+ "loss": 0.1597,
878
+ "step": 1420
879
+ },
880
+ {
881
+ "epoch": 2.84,
882
+ "learning_rate": 4.326043737574553e-05,
883
+ "loss": 0.1476,
884
+ "step": 1430
885
+ },
886
+ {
887
+ "epoch": 2.86,
888
+ "learning_rate": 4.286282306163022e-05,
889
+ "loss": 0.1634,
890
+ "step": 1440
891
+ },
892
+ {
893
+ "epoch": 2.88,
894
+ "learning_rate": 4.2465208747514914e-05,
895
+ "loss": 0.3117,
896
+ "step": 1450
897
+ },
898
+ {
899
+ "epoch": 2.9,
900
+ "learning_rate": 4.206759443339961e-05,
901
+ "loss": 0.1822,
902
+ "step": 1460
903
+ },
904
+ {
905
+ "epoch": 2.92,
906
+ "learning_rate": 4.1669980119284293e-05,
907
+ "loss": 0.2559,
908
+ "step": 1470
909
+ },
910
+ {
911
+ "epoch": 2.94,
912
+ "learning_rate": 4.1272365805168987e-05,
913
+ "loss": 0.0924,
914
+ "step": 1480
915
+ },
916
+ {
917
+ "epoch": 2.96,
918
+ "learning_rate": 4.087475149105368e-05,
919
+ "loss": 0.095,
920
+ "step": 1490
921
+ },
922
+ {
923
+ "epoch": 2.98,
924
+ "learning_rate": 4.047713717693837e-05,
925
+ "loss": 0.0947,
926
+ "step": 1500
927
+ },
928
+ {
929
+ "epoch": 2.98,
930
+ "eval_accuracy": 0.9305555820465088,
931
+ "eval_loss": 0.23964740335941315,
932
+ "eval_runtime": 18.9345,
933
+ "eval_samples_per_second": 26.618,
934
+ "eval_steps_per_second": 6.655,
935
+ "step": 1500
936
+ },
937
+ {
938
+ "epoch": 3.0,
939
+ "learning_rate": 4.0079522862823066e-05,
940
+ "loss": 0.1292,
941
+ "step": 1510
942
+ },
943
+ {
944
+ "epoch": 3.02,
945
+ "learning_rate": 3.968190854870776e-05,
946
+ "loss": 0.0505,
947
+ "step": 1520
948
+ },
949
+ {
950
+ "epoch": 3.04,
951
+ "learning_rate": 3.9284294234592445e-05,
952
+ "loss": 0.0629,
953
+ "step": 1530
954
+ },
955
+ {
956
+ "epoch": 3.06,
957
+ "learning_rate": 3.888667992047714e-05,
958
+ "loss": 0.2428,
959
+ "step": 1540
960
+ },
961
+ {
962
+ "epoch": 3.08,
963
+ "learning_rate": 3.848906560636183e-05,
964
+ "loss": 0.0792,
965
+ "step": 1550
966
+ },
967
+ {
968
+ "epoch": 3.1,
969
+ "learning_rate": 3.8091451292246525e-05,
970
+ "loss": 0.1127,
971
+ "step": 1560
972
+ },
973
+ {
974
+ "epoch": 3.12,
975
+ "learning_rate": 3.769383697813121e-05,
976
+ "loss": 0.0928,
977
+ "step": 1570
978
+ },
979
+ {
980
+ "epoch": 3.14,
981
+ "learning_rate": 3.7296222664015904e-05,
982
+ "loss": 0.1148,
983
+ "step": 1580
984
+ },
985
+ {
986
+ "epoch": 3.16,
987
+ "learning_rate": 3.68986083499006e-05,
988
+ "loss": 0.1969,
989
+ "step": 1590
990
+ },
991
+ {
992
+ "epoch": 3.18,
993
+ "learning_rate": 3.650099403578529e-05,
994
+ "loss": 0.0891,
995
+ "step": 1600
996
+ },
997
+ {
998
+ "epoch": 3.2,
999
+ "learning_rate": 3.6103379721669984e-05,
1000
+ "loss": 0.1074,
1001
+ "step": 1610
1002
+ },
1003
+ {
1004
+ "epoch": 3.22,
1005
+ "learning_rate": 3.570576540755468e-05,
1006
+ "loss": 0.1484,
1007
+ "step": 1620
1008
+ },
1009
+ {
1010
+ "epoch": 3.24,
1011
+ "learning_rate": 3.530815109343937e-05,
1012
+ "loss": 0.2352,
1013
+ "step": 1630
1014
+ },
1015
+ {
1016
+ "epoch": 3.26,
1017
+ "learning_rate": 3.491053677932406e-05,
1018
+ "loss": 0.0469,
1019
+ "step": 1640
1020
+ },
1021
+ {
1022
+ "epoch": 3.28,
1023
+ "learning_rate": 3.451292246520875e-05,
1024
+ "loss": 0.1484,
1025
+ "step": 1650
1026
+ },
1027
+ {
1028
+ "epoch": 3.3,
1029
+ "learning_rate": 3.4115308151093436e-05,
1030
+ "loss": 0.1446,
1031
+ "step": 1660
1032
+ },
1033
+ {
1034
+ "epoch": 3.32,
1035
+ "learning_rate": 3.371769383697813e-05,
1036
+ "loss": 0.0165,
1037
+ "step": 1670
1038
+ },
1039
+ {
1040
+ "epoch": 3.34,
1041
+ "learning_rate": 3.332007952286282e-05,
1042
+ "loss": 0.0797,
1043
+ "step": 1680
1044
+ },
1045
+ {
1046
+ "epoch": 3.36,
1047
+ "learning_rate": 3.2922465208747515e-05,
1048
+ "loss": 0.1391,
1049
+ "step": 1690
1050
+ },
1051
+ {
1052
+ "epoch": 3.38,
1053
+ "learning_rate": 3.252485089463221e-05,
1054
+ "loss": 0.0918,
1055
+ "step": 1700
1056
+ },
1057
+ {
1058
+ "epoch": 3.4,
1059
+ "learning_rate": 3.21272365805169e-05,
1060
+ "loss": 0.1694,
1061
+ "step": 1710
1062
+ },
1063
+ {
1064
+ "epoch": 3.42,
1065
+ "learning_rate": 3.1729622266401595e-05,
1066
+ "loss": 0.1379,
1067
+ "step": 1720
1068
+ },
1069
+ {
1070
+ "epoch": 3.44,
1071
+ "learning_rate": 3.133200795228629e-05,
1072
+ "loss": 0.1016,
1073
+ "step": 1730
1074
+ },
1075
+ {
1076
+ "epoch": 3.46,
1077
+ "learning_rate": 3.0934393638170974e-05,
1078
+ "loss": 0.0449,
1079
+ "step": 1740
1080
+ },
1081
+ {
1082
+ "epoch": 3.48,
1083
+ "learning_rate": 3.053677932405567e-05,
1084
+ "loss": 0.0124,
1085
+ "step": 1750
1086
+ },
1087
+ {
1088
+ "epoch": 3.5,
1089
+ "learning_rate": 3.0139165009940357e-05,
1090
+ "loss": 0.0186,
1091
+ "step": 1760
1092
+ },
1093
+ {
1094
+ "epoch": 3.52,
1095
+ "learning_rate": 2.974155069582505e-05,
1096
+ "loss": 0.1103,
1097
+ "step": 1770
1098
+ },
1099
+ {
1100
+ "epoch": 3.54,
1101
+ "learning_rate": 2.934393638170974e-05,
1102
+ "loss": 0.063,
1103
+ "step": 1780
1104
+ },
1105
+ {
1106
+ "epoch": 3.56,
1107
+ "learning_rate": 2.8946322067594433e-05,
1108
+ "loss": 0.061,
1109
+ "step": 1790
1110
+ },
1111
+ {
1112
+ "epoch": 3.58,
1113
+ "learning_rate": 2.8548707753479126e-05,
1114
+ "loss": 0.1044,
1115
+ "step": 1800
1116
+ },
1117
+ {
1118
+ "epoch": 3.6,
1119
+ "learning_rate": 2.815109343936382e-05,
1120
+ "loss": 0.1516,
1121
+ "step": 1810
1122
+ },
1123
+ {
1124
+ "epoch": 3.62,
1125
+ "learning_rate": 2.775347912524851e-05,
1126
+ "loss": 0.1427,
1127
+ "step": 1820
1128
+ },
1129
+ {
1130
+ "epoch": 3.64,
1131
+ "learning_rate": 2.7355864811133202e-05,
1132
+ "loss": 0.0448,
1133
+ "step": 1830
1134
+ },
1135
+ {
1136
+ "epoch": 3.66,
1137
+ "learning_rate": 2.6958250497017895e-05,
1138
+ "loss": 0.0703,
1139
+ "step": 1840
1140
+ },
1141
+ {
1142
+ "epoch": 3.68,
1143
+ "learning_rate": 2.656063618290259e-05,
1144
+ "loss": 0.0905,
1145
+ "step": 1850
1146
+ },
1147
+ {
1148
+ "epoch": 3.7,
1149
+ "learning_rate": 2.616302186878728e-05,
1150
+ "loss": 0.081,
1151
+ "step": 1860
1152
+ },
1153
+ {
1154
+ "epoch": 3.72,
1155
+ "learning_rate": 2.576540755467197e-05,
1156
+ "loss": 0.1642,
1157
+ "step": 1870
1158
+ },
1159
+ {
1160
+ "epoch": 3.74,
1161
+ "learning_rate": 2.5367793240556658e-05,
1162
+ "loss": 0.1915,
1163
+ "step": 1880
1164
+ },
1165
+ {
1166
+ "epoch": 3.76,
1167
+ "learning_rate": 2.4970178926441354e-05,
1168
+ "loss": 0.0696,
1169
+ "step": 1890
1170
+ },
1171
+ {
1172
+ "epoch": 3.78,
1173
+ "learning_rate": 2.4572564612326048e-05,
1174
+ "loss": 0.0605,
1175
+ "step": 1900
1176
+ },
1177
+ {
1178
+ "epoch": 3.8,
1179
+ "learning_rate": 2.4174950298210737e-05,
1180
+ "loss": 0.1562,
1181
+ "step": 1910
1182
+ },
1183
+ {
1184
+ "epoch": 3.82,
1185
+ "learning_rate": 2.3777335984095427e-05,
1186
+ "loss": 0.1316,
1187
+ "step": 1920
1188
+ },
1189
+ {
1190
+ "epoch": 3.84,
1191
+ "learning_rate": 2.337972166998012e-05,
1192
+ "loss": 0.1364,
1193
+ "step": 1930
1194
+ },
1195
+ {
1196
+ "epoch": 3.86,
1197
+ "learning_rate": 2.2982107355864813e-05,
1198
+ "loss": 0.0766,
1199
+ "step": 1940
1200
+ },
1201
+ {
1202
+ "epoch": 3.88,
1203
+ "learning_rate": 2.2584493041749506e-05,
1204
+ "loss": 0.2037,
1205
+ "step": 1950
1206
+ },
1207
+ {
1208
+ "epoch": 3.9,
1209
+ "learning_rate": 2.2186878727634196e-05,
1210
+ "loss": 0.029,
1211
+ "step": 1960
1212
+ },
1213
+ {
1214
+ "epoch": 3.92,
1215
+ "learning_rate": 2.1789264413518886e-05,
1216
+ "loss": 0.0443,
1217
+ "step": 1970
1218
+ },
1219
+ {
1220
+ "epoch": 3.94,
1221
+ "learning_rate": 2.139165009940358e-05,
1222
+ "loss": 0.0494,
1223
+ "step": 1980
1224
+ },
1225
+ {
1226
+ "epoch": 3.96,
1227
+ "learning_rate": 2.0994035785288272e-05,
1228
+ "loss": 0.0766,
1229
+ "step": 1990
1230
+ },
1231
+ {
1232
+ "epoch": 3.98,
1233
+ "learning_rate": 2.0596421471172962e-05,
1234
+ "loss": 0.1602,
1235
+ "step": 2000
1236
+ },
1237
+ {
1238
+ "epoch": 3.98,
1239
+ "eval_accuracy": 0.9444444179534912,
1240
+ "eval_loss": 0.26106151938438416,
1241
+ "eval_runtime": 19.5055,
1242
+ "eval_samples_per_second": 25.839,
1243
+ "eval_steps_per_second": 6.46,
1244
+ "step": 2000
1245
+ },
1246
+ {
1247
+ "epoch": 4.0,
1248
+ "learning_rate": 2.0198807157057655e-05,
1249
+ "loss": 0.0202,
1250
+ "step": 2010
1251
+ },
1252
+ {
1253
+ "epoch": 4.02,
1254
+ "learning_rate": 1.9801192842942348e-05,
1255
+ "loss": 0.0856,
1256
+ "step": 2020
1257
+ },
1258
+ {
1259
+ "epoch": 4.04,
1260
+ "learning_rate": 1.9403578528827038e-05,
1261
+ "loss": 0.0999,
1262
+ "step": 2030
1263
+ },
1264
+ {
1265
+ "epoch": 4.06,
1266
+ "learning_rate": 1.900596421471173e-05,
1267
+ "loss": 0.0053,
1268
+ "step": 2040
1269
+ },
1270
+ {
1271
+ "epoch": 4.08,
1272
+ "learning_rate": 1.860834990059642e-05,
1273
+ "loss": 0.1129,
1274
+ "step": 2050
1275
+ },
1276
+ {
1277
+ "epoch": 4.1,
1278
+ "learning_rate": 1.8210735586481114e-05,
1279
+ "loss": 0.0083,
1280
+ "step": 2060
1281
+ },
1282
+ {
1283
+ "epoch": 4.12,
1284
+ "learning_rate": 1.7813121272365807e-05,
1285
+ "loss": 0.0857,
1286
+ "step": 2070
1287
+ },
1288
+ {
1289
+ "epoch": 4.14,
1290
+ "learning_rate": 1.7415506958250497e-05,
1291
+ "loss": 0.0475,
1292
+ "step": 2080
1293
+ },
1294
+ {
1295
+ "epoch": 4.15,
1296
+ "learning_rate": 1.701789264413519e-05,
1297
+ "loss": 0.0267,
1298
+ "step": 2090
1299
+ },
1300
+ {
1301
+ "epoch": 4.17,
1302
+ "learning_rate": 1.662027833001988e-05,
1303
+ "loss": 0.0469,
1304
+ "step": 2100
1305
+ },
1306
+ {
1307
+ "epoch": 4.19,
1308
+ "learning_rate": 1.6222664015904573e-05,
1309
+ "loss": 0.0346,
1310
+ "step": 2110
1311
+ },
1312
+ {
1313
+ "epoch": 4.21,
1314
+ "learning_rate": 1.5825049701789266e-05,
1315
+ "loss": 0.0822,
1316
+ "step": 2120
1317
+ },
1318
+ {
1319
+ "epoch": 4.23,
1320
+ "learning_rate": 1.542743538767396e-05,
1321
+ "loss": 0.0803,
1322
+ "step": 2130
1323
+ },
1324
+ {
1325
+ "epoch": 4.25,
1326
+ "learning_rate": 1.5029821073558647e-05,
1327
+ "loss": 0.0077,
1328
+ "step": 2140
1329
+ },
1330
+ {
1331
+ "epoch": 4.27,
1332
+ "learning_rate": 1.463220675944334e-05,
1333
+ "loss": 0.0347,
1334
+ "step": 2150
1335
+ },
1336
+ {
1337
+ "epoch": 4.29,
1338
+ "learning_rate": 1.4234592445328032e-05,
1339
+ "loss": 0.0665,
1340
+ "step": 2160
1341
+ },
1342
+ {
1343
+ "epoch": 4.31,
1344
+ "learning_rate": 1.3836978131212725e-05,
1345
+ "loss": 0.0779,
1346
+ "step": 2170
1347
+ },
1348
+ {
1349
+ "epoch": 4.33,
1350
+ "learning_rate": 1.3439363817097416e-05,
1351
+ "loss": 0.056,
1352
+ "step": 2180
1353
+ },
1354
+ {
1355
+ "epoch": 4.35,
1356
+ "learning_rate": 1.304174950298211e-05,
1357
+ "loss": 0.1503,
1358
+ "step": 2190
1359
+ },
1360
+ {
1361
+ "epoch": 4.37,
1362
+ "learning_rate": 1.26441351888668e-05,
1363
+ "loss": 0.0032,
1364
+ "step": 2200
1365
+ },
1366
+ {
1367
+ "epoch": 4.39,
1368
+ "learning_rate": 1.2246520874751492e-05,
1369
+ "loss": 0.0457,
1370
+ "step": 2210
1371
+ },
1372
+ {
1373
+ "epoch": 4.41,
1374
+ "learning_rate": 1.1848906560636184e-05,
1375
+ "loss": 0.0077,
1376
+ "step": 2220
1377
+ },
1378
+ {
1379
+ "epoch": 4.43,
1380
+ "learning_rate": 1.1451292246520875e-05,
1381
+ "loss": 0.1558,
1382
+ "step": 2230
1383
+ },
1384
+ {
1385
+ "epoch": 4.45,
1386
+ "learning_rate": 1.1053677932405568e-05,
1387
+ "loss": 0.0307,
1388
+ "step": 2240
1389
+ },
1390
+ {
1391
+ "epoch": 4.47,
1392
+ "learning_rate": 1.0656063618290258e-05,
1393
+ "loss": 0.1193,
1394
+ "step": 2250
1395
+ },
1396
+ {
1397
+ "epoch": 4.49,
1398
+ "learning_rate": 1.0258449304174951e-05,
1399
+ "loss": 0.0638,
1400
+ "step": 2260
1401
+ },
1402
+ {
1403
+ "epoch": 4.51,
1404
+ "learning_rate": 9.860834990059643e-06,
1405
+ "loss": 0.059,
1406
+ "step": 2270
1407
+ },
1408
+ {
1409
+ "epoch": 4.53,
1410
+ "learning_rate": 9.463220675944334e-06,
1411
+ "loss": 0.0875,
1412
+ "step": 2280
1413
+ },
1414
+ {
1415
+ "epoch": 4.55,
1416
+ "learning_rate": 9.065606361829027e-06,
1417
+ "loss": 0.1511,
1418
+ "step": 2290
1419
+ },
1420
+ {
1421
+ "epoch": 4.57,
1422
+ "learning_rate": 8.667992047713717e-06,
1423
+ "loss": 0.1217,
1424
+ "step": 2300
1425
+ },
1426
+ {
1427
+ "epoch": 4.59,
1428
+ "learning_rate": 8.27037773359841e-06,
1429
+ "loss": 0.2091,
1430
+ "step": 2310
1431
+ },
1432
+ {
1433
+ "epoch": 4.61,
1434
+ "learning_rate": 7.872763419483102e-06,
1435
+ "loss": 0.1723,
1436
+ "step": 2320
1437
+ },
1438
+ {
1439
+ "epoch": 4.63,
1440
+ "learning_rate": 7.475149105367793e-06,
1441
+ "loss": 0.0763,
1442
+ "step": 2330
1443
+ },
1444
+ {
1445
+ "epoch": 4.65,
1446
+ "learning_rate": 7.077534791252485e-06,
1447
+ "loss": 0.0172,
1448
+ "step": 2340
1449
+ },
1450
+ {
1451
+ "epoch": 4.67,
1452
+ "learning_rate": 6.679920477137178e-06,
1453
+ "loss": 0.0352,
1454
+ "step": 2350
1455
+ },
1456
+ {
1457
+ "epoch": 4.69,
1458
+ "learning_rate": 6.282306163021868e-06,
1459
+ "loss": 0.1989,
1460
+ "step": 2360
1461
+ },
1462
+ {
1463
+ "epoch": 4.71,
1464
+ "learning_rate": 5.8846918489065606e-06,
1465
+ "loss": 0.0896,
1466
+ "step": 2370
1467
+ },
1468
+ {
1469
+ "epoch": 4.73,
1470
+ "learning_rate": 5.487077534791253e-06,
1471
+ "loss": 0.0728,
1472
+ "step": 2380
1473
+ },
1474
+ {
1475
+ "epoch": 4.75,
1476
+ "learning_rate": 5.089463220675945e-06,
1477
+ "loss": 0.0482,
1478
+ "step": 2390
1479
+ },
1480
+ {
1481
+ "epoch": 4.77,
1482
+ "learning_rate": 4.6918489065606366e-06,
1483
+ "loss": 0.1288,
1484
+ "step": 2400
1485
+ },
1486
+ {
1487
+ "epoch": 4.79,
1488
+ "learning_rate": 4.294234592445328e-06,
1489
+ "loss": 0.0756,
1490
+ "step": 2410
1491
+ },
1492
+ {
1493
+ "epoch": 4.81,
1494
+ "learning_rate": 3.89662027833002e-06,
1495
+ "loss": 0.051,
1496
+ "step": 2420
1497
+ },
1498
+ {
1499
+ "epoch": 4.83,
1500
+ "learning_rate": 3.4990059642147117e-06,
1501
+ "loss": 0.0874,
1502
+ "step": 2430
1503
+ },
1504
+ {
1505
+ "epoch": 4.85,
1506
+ "learning_rate": 3.101391650099404e-06,
1507
+ "loss": 0.0381,
1508
+ "step": 2440
1509
+ },
1510
+ {
1511
+ "epoch": 4.87,
1512
+ "learning_rate": 2.7037773359840955e-06,
1513
+ "loss": 0.0031,
1514
+ "step": 2450
1515
+ },
1516
+ {
1517
+ "epoch": 4.89,
1518
+ "learning_rate": 2.3061630218687873e-06,
1519
+ "loss": 0.0371,
1520
+ "step": 2460
1521
+ },
1522
+ {
1523
+ "epoch": 4.91,
1524
+ "learning_rate": 1.9085487077534792e-06,
1525
+ "loss": 0.007,
1526
+ "step": 2470
1527
+ },
1528
+ {
1529
+ "epoch": 4.93,
1530
+ "learning_rate": 1.510934393638171e-06,
1531
+ "loss": 0.115,
1532
+ "step": 2480
1533
+ },
1534
+ {
1535
+ "epoch": 4.95,
1536
+ "learning_rate": 1.113320079522863e-06,
1537
+ "loss": 0.0492,
1538
+ "step": 2490
1539
+ },
1540
+ {
1541
+ "epoch": 4.97,
1542
+ "learning_rate": 7.157057654075547e-07,
1543
+ "loss": 0.0104,
1544
+ "step": 2500
1545
+ },
1546
+ {
1547
+ "epoch": 4.97,
1548
+ "eval_accuracy": 0.9484127163887024,
1549
+ "eval_loss": 0.2534567415714264,
1550
+ "eval_runtime": 18.9048,
1551
+ "eval_samples_per_second": 26.66,
1552
+ "eval_steps_per_second": 6.665,
1553
+ "step": 2500
1554
+ },
1555
+ {
1556
+ "epoch": 4.99,
1557
+ "learning_rate": 3.1809145129224657e-07,
1558
+ "loss": 0.0589,
1559
+ "step": 2510
1560
+ }
1561
+ ],
1562
+ "max_steps": 2515,
1563
+ "num_train_epochs": 5,
1564
+ "total_flos": 7.174884566087474e+17,
1565
+ "trial_name": null,
1566
+ "trial_params": null
1567
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f1ad8e035674c9486798c9d1ff63ff038ab967349d5c2e0200e5cea5884540bf
3
+ size 3643