user commited on
Commit
8c00a8e
1 Parent(s): 4d4ec77
adapter_config.json ADDED
@@ -0,0 +1,18 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name_or_path": "./llama-13b-hf",
3
+ "bias": "none",
4
+ "enable_lora": null,
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "lora_alpha": 16,
8
+ "lora_dropout": 0.05,
9
+ "merge_weights": false,
10
+ "modules_to_save": null,
11
+ "peft_type": "LORA",
12
+ "r": 16,
13
+ "target_modules": [
14
+ "q_proj",
15
+ "v_proj"
16
+ ],
17
+ "task_type": "CAUSAL_LM"
18
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:72917740bc80a26c29e9a02478fc5ea55c48b551a05eefae56e46be131dd68ba
3
+ size 52486477
checkpoint-1800/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87be2545cbc8112710f6f65d2b14e3a73a3ab35e030f0b1683aa2595d3324a91
3
+ size 104992197
checkpoint-1800/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3a750edb8a02cc6e524dbe112836640a94876a87b3fac17bcbca0dc9911670bc
3
+ size 52486477
checkpoint-1800/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f92f71754c3ab8dcc145c29ecbc12dc56481feeb48df842033738413f6e05877
3
+ size 14583
checkpoint-1800/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:59bf234fec84aaa8cf5a7a208b2e2ef7029d6e06152c2c16e0c47a61f0c876e1
3
+ size 14583
checkpoint-1800/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:00c8e5257dc04b045bc4088585391569697ae60fe73e1b5b4f0bdec0a1f699ae
3
+ size 14583
checkpoint-1800/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a8aae3129e012e217528b68b83a8f8650d86fd524dce4c38b51add95b7fa5ff7
3
+ size 14583
checkpoint-1800/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:31416101e0bd09029cd59a809386a2ae447140e48a6d9f4c193fa2f202f28b2b
3
+ size 557
checkpoint-1800/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f02b3252c69b5f71ddb00d226117ca798ba7a3677be5f1343a198acb62523b82
3
+ size 627
checkpoint-1800/trainer_state.json ADDED
@@ -0,0 +1,1168 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9558648467063904,
3
+ "best_model_checkpoint": "./lora-alpaca/checkpoint-1800",
4
+ "epoch": 2.2886204704386524,
5
+ "global_step": 1800,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 2.6999999999999996e-05,
13
+ "loss": 1.6482,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.03,
18
+ "learning_rate": 5.399999999999999e-05,
19
+ "loss": 1.5884,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.04,
24
+ "learning_rate": 8.4e-05,
25
+ "loss": 1.5166,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.05,
30
+ "learning_rate": 0.00011399999999999999,
31
+ "loss": 1.3197,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.06,
36
+ "learning_rate": 0.00014399999999999998,
37
+ "loss": 1.1732,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.08,
42
+ "learning_rate": 0.00017399999999999997,
43
+ "loss": 1.1358,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.09,
48
+ "learning_rate": 0.000204,
49
+ "loss": 1.0792,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.1,
54
+ "learning_rate": 0.000234,
55
+ "loss": 1.062,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.11,
60
+ "learning_rate": 0.00026399999999999997,
61
+ "loss": 1.0665,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.13,
66
+ "learning_rate": 0.000294,
67
+ "loss": 1.0641,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.14,
72
+ "learning_rate": 0.00029893711248892826,
73
+ "loss": 1.0449,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.15,
78
+ "learning_rate": 0.00029760850310008854,
79
+ "loss": 1.0356,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.17,
84
+ "learning_rate": 0.0002962798937112489,
85
+ "loss": 1.0537,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.18,
90
+ "learning_rate": 0.0002949512843224092,
91
+ "loss": 1.0285,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.19,
96
+ "learning_rate": 0.0002936226749335695,
97
+ "loss": 1.0197,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.2,
102
+ "learning_rate": 0.0002922940655447298,
103
+ "loss": 1.0171,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.22,
108
+ "learning_rate": 0.00029096545615589015,
109
+ "loss": 1.0112,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.23,
114
+ "learning_rate": 0.0002896368467670505,
115
+ "loss": 1.0299,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.24,
120
+ "learning_rate": 0.00028830823737821076,
121
+ "loss": 1.0243,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.25,
126
+ "learning_rate": 0.0002869796279893711,
127
+ "loss": 1.0118,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.25,
132
+ "eval_loss": 1.0055800676345825,
133
+ "eval_runtime": 82.8279,
134
+ "eval_samples_per_second": 24.146,
135
+ "eval_steps_per_second": 0.761,
136
+ "step": 200
137
+ },
138
+ {
139
+ "epoch": 0.27,
140
+ "learning_rate": 0.00028565101860053143,
141
+ "loss": 0.9956,
142
+ "step": 210
143
+ },
144
+ {
145
+ "epoch": 0.28,
146
+ "learning_rate": 0.00028432240921169176,
147
+ "loss": 1.0039,
148
+ "step": 220
149
+ },
150
+ {
151
+ "epoch": 0.29,
152
+ "learning_rate": 0.00028299379982285204,
153
+ "loss": 1.0026,
154
+ "step": 230
155
+ },
156
+ {
157
+ "epoch": 0.31,
158
+ "learning_rate": 0.00028166519043401237,
159
+ "loss": 1.0148,
160
+ "step": 240
161
+ },
162
+ {
163
+ "epoch": 0.32,
164
+ "learning_rate": 0.0002803365810451727,
165
+ "loss": 1.0065,
166
+ "step": 250
167
+ },
168
+ {
169
+ "epoch": 0.33,
170
+ "learning_rate": 0.00027900797165633304,
171
+ "loss": 0.9965,
172
+ "step": 260
173
+ },
174
+ {
175
+ "epoch": 0.34,
176
+ "learning_rate": 0.0002776793622674933,
177
+ "loss": 1.0037,
178
+ "step": 270
179
+ },
180
+ {
181
+ "epoch": 0.36,
182
+ "learning_rate": 0.00027635075287865365,
183
+ "loss": 1.0149,
184
+ "step": 280
185
+ },
186
+ {
187
+ "epoch": 0.37,
188
+ "learning_rate": 0.000275022143489814,
189
+ "loss": 1.0136,
190
+ "step": 290
191
+ },
192
+ {
193
+ "epoch": 0.38,
194
+ "learning_rate": 0.00027369353410097426,
195
+ "loss": 0.9937,
196
+ "step": 300
197
+ },
198
+ {
199
+ "epoch": 0.39,
200
+ "learning_rate": 0.0002723649247121346,
201
+ "loss": 0.987,
202
+ "step": 310
203
+ },
204
+ {
205
+ "epoch": 0.41,
206
+ "learning_rate": 0.0002710363153232949,
207
+ "loss": 1.016,
208
+ "step": 320
209
+ },
210
+ {
211
+ "epoch": 0.42,
212
+ "learning_rate": 0.00026970770593445526,
213
+ "loss": 1.0136,
214
+ "step": 330
215
+ },
216
+ {
217
+ "epoch": 0.43,
218
+ "learning_rate": 0.00026837909654561554,
219
+ "loss": 0.9926,
220
+ "step": 340
221
+ },
222
+ {
223
+ "epoch": 0.45,
224
+ "learning_rate": 0.00026705048715677587,
225
+ "loss": 0.9922,
226
+ "step": 350
227
+ },
228
+ {
229
+ "epoch": 0.46,
230
+ "learning_rate": 0.0002657218777679362,
231
+ "loss": 0.9883,
232
+ "step": 360
233
+ },
234
+ {
235
+ "epoch": 0.47,
236
+ "learning_rate": 0.00026439326837909654,
237
+ "loss": 1.0076,
238
+ "step": 370
239
+ },
240
+ {
241
+ "epoch": 0.48,
242
+ "learning_rate": 0.0002630646589902568,
243
+ "loss": 0.9873,
244
+ "step": 380
245
+ },
246
+ {
247
+ "epoch": 0.5,
248
+ "learning_rate": 0.0002617360496014172,
249
+ "loss": 0.988,
250
+ "step": 390
251
+ },
252
+ {
253
+ "epoch": 0.51,
254
+ "learning_rate": 0.0002604074402125775,
255
+ "loss": 0.9902,
256
+ "step": 400
257
+ },
258
+ {
259
+ "epoch": 0.51,
260
+ "eval_loss": 0.9863091111183167,
261
+ "eval_runtime": 82.9481,
262
+ "eval_samples_per_second": 24.111,
263
+ "eval_steps_per_second": 0.76,
264
+ "step": 400
265
+ },
266
+ {
267
+ "epoch": 0.52,
268
+ "learning_rate": 0.00025907883082373776,
269
+ "loss": 0.9966,
270
+ "step": 410
271
+ },
272
+ {
273
+ "epoch": 0.53,
274
+ "learning_rate": 0.00025775022143489815,
275
+ "loss": 0.9914,
276
+ "step": 420
277
+ },
278
+ {
279
+ "epoch": 0.55,
280
+ "learning_rate": 0.0002564216120460584,
281
+ "loss": 0.9924,
282
+ "step": 430
283
+ },
284
+ {
285
+ "epoch": 0.56,
286
+ "learning_rate": 0.00025509300265721876,
287
+ "loss": 0.9948,
288
+ "step": 440
289
+ },
290
+ {
291
+ "epoch": 0.57,
292
+ "learning_rate": 0.0002537643932683791,
293
+ "loss": 0.9879,
294
+ "step": 450
295
+ },
296
+ {
297
+ "epoch": 0.58,
298
+ "learning_rate": 0.0002524357838795394,
299
+ "loss": 0.988,
300
+ "step": 460
301
+ },
302
+ {
303
+ "epoch": 0.6,
304
+ "learning_rate": 0.0002511071744906997,
305
+ "loss": 0.9928,
306
+ "step": 470
307
+ },
308
+ {
309
+ "epoch": 0.61,
310
+ "learning_rate": 0.00024977856510186003,
311
+ "loss": 0.9916,
312
+ "step": 480
313
+ },
314
+ {
315
+ "epoch": 0.62,
316
+ "learning_rate": 0.00024844995571302037,
317
+ "loss": 0.994,
318
+ "step": 490
319
+ },
320
+ {
321
+ "epoch": 0.64,
322
+ "learning_rate": 0.0002471213463241807,
323
+ "loss": 0.996,
324
+ "step": 500
325
+ },
326
+ {
327
+ "epoch": 0.65,
328
+ "learning_rate": 0.000245792736935341,
329
+ "loss": 0.9783,
330
+ "step": 510
331
+ },
332
+ {
333
+ "epoch": 0.66,
334
+ "learning_rate": 0.0002444641275465013,
335
+ "loss": 0.9798,
336
+ "step": 520
337
+ },
338
+ {
339
+ "epoch": 0.67,
340
+ "learning_rate": 0.00024313551815766162,
341
+ "loss": 0.9791,
342
+ "step": 530
343
+ },
344
+ {
345
+ "epoch": 0.69,
346
+ "learning_rate": 0.00024180690876882195,
347
+ "loss": 0.9894,
348
+ "step": 540
349
+ },
350
+ {
351
+ "epoch": 0.7,
352
+ "learning_rate": 0.00024047829937998225,
353
+ "loss": 0.9904,
354
+ "step": 550
355
+ },
356
+ {
357
+ "epoch": 0.71,
358
+ "learning_rate": 0.00023914968999114256,
359
+ "loss": 0.9942,
360
+ "step": 560
361
+ },
362
+ {
363
+ "epoch": 0.72,
364
+ "learning_rate": 0.0002378210806023029,
365
+ "loss": 0.9897,
366
+ "step": 570
367
+ },
368
+ {
369
+ "epoch": 0.74,
370
+ "learning_rate": 0.0002364924712134632,
371
+ "loss": 0.9741,
372
+ "step": 580
373
+ },
374
+ {
375
+ "epoch": 0.75,
376
+ "learning_rate": 0.00023516386182462353,
377
+ "loss": 0.9631,
378
+ "step": 590
379
+ },
380
+ {
381
+ "epoch": 0.76,
382
+ "learning_rate": 0.00023383525243578384,
383
+ "loss": 0.9784,
384
+ "step": 600
385
+ },
386
+ {
387
+ "epoch": 0.76,
388
+ "eval_loss": 0.9769166111946106,
389
+ "eval_runtime": 78.8243,
390
+ "eval_samples_per_second": 25.373,
391
+ "eval_steps_per_second": 0.799,
392
+ "step": 600
393
+ },
394
+ {
395
+ "epoch": 0.78,
396
+ "learning_rate": 0.0002325066430469442,
397
+ "loss": 0.9703,
398
+ "step": 610
399
+ },
400
+ {
401
+ "epoch": 0.79,
402
+ "learning_rate": 0.00023117803365810448,
403
+ "loss": 0.9914,
404
+ "step": 620
405
+ },
406
+ {
407
+ "epoch": 0.8,
408
+ "learning_rate": 0.00022984942426926484,
409
+ "loss": 0.9522,
410
+ "step": 630
411
+ },
412
+ {
413
+ "epoch": 0.81,
414
+ "learning_rate": 0.00022852081488042514,
415
+ "loss": 0.9773,
416
+ "step": 640
417
+ },
418
+ {
419
+ "epoch": 0.83,
420
+ "learning_rate": 0.00022719220549158547,
421
+ "loss": 0.9954,
422
+ "step": 650
423
+ },
424
+ {
425
+ "epoch": 0.84,
426
+ "learning_rate": 0.00022586359610274578,
427
+ "loss": 0.9768,
428
+ "step": 660
429
+ },
430
+ {
431
+ "epoch": 0.85,
432
+ "learning_rate": 0.0002245349867139061,
433
+ "loss": 0.9908,
434
+ "step": 670
435
+ },
436
+ {
437
+ "epoch": 0.86,
438
+ "learning_rate": 0.00022320637732506642,
439
+ "loss": 0.9944,
440
+ "step": 680
441
+ },
442
+ {
443
+ "epoch": 0.88,
444
+ "learning_rate": 0.00022187776793622672,
445
+ "loss": 0.9772,
446
+ "step": 690
447
+ },
448
+ {
449
+ "epoch": 0.89,
450
+ "learning_rate": 0.00022054915854738706,
451
+ "loss": 0.9697,
452
+ "step": 700
453
+ },
454
+ {
455
+ "epoch": 0.9,
456
+ "learning_rate": 0.00021922054915854736,
457
+ "loss": 0.9652,
458
+ "step": 710
459
+ },
460
+ {
461
+ "epoch": 0.92,
462
+ "learning_rate": 0.0002178919397697077,
463
+ "loss": 0.9738,
464
+ "step": 720
465
+ },
466
+ {
467
+ "epoch": 0.93,
468
+ "learning_rate": 0.000216563330380868,
469
+ "loss": 0.9892,
470
+ "step": 730
471
+ },
472
+ {
473
+ "epoch": 0.94,
474
+ "learning_rate": 0.00021523472099202833,
475
+ "loss": 0.9731,
476
+ "step": 740
477
+ },
478
+ {
479
+ "epoch": 0.95,
480
+ "learning_rate": 0.00021390611160318864,
481
+ "loss": 0.985,
482
+ "step": 750
483
+ },
484
+ {
485
+ "epoch": 0.97,
486
+ "learning_rate": 0.00021257750221434897,
487
+ "loss": 0.9918,
488
+ "step": 760
489
+ },
490
+ {
491
+ "epoch": 0.98,
492
+ "learning_rate": 0.00021124889282550928,
493
+ "loss": 0.9603,
494
+ "step": 770
495
+ },
496
+ {
497
+ "epoch": 0.99,
498
+ "learning_rate": 0.0002099202834366696,
499
+ "loss": 0.9686,
500
+ "step": 780
501
+ },
502
+ {
503
+ "epoch": 1.0,
504
+ "learning_rate": 0.00020859167404782992,
505
+ "loss": 0.9761,
506
+ "step": 790
507
+ },
508
+ {
509
+ "epoch": 1.02,
510
+ "learning_rate": 0.00020726306465899025,
511
+ "loss": 0.9812,
512
+ "step": 800
513
+ },
514
+ {
515
+ "epoch": 1.02,
516
+ "eval_loss": 0.970382571220398,
517
+ "eval_runtime": 78.9078,
518
+ "eval_samples_per_second": 25.346,
519
+ "eval_steps_per_second": 0.798,
520
+ "step": 800
521
+ },
522
+ {
523
+ "epoch": 1.03,
524
+ "learning_rate": 0.00020593445527015056,
525
+ "loss": 0.95,
526
+ "step": 810
527
+ },
528
+ {
529
+ "epoch": 1.04,
530
+ "learning_rate": 0.0002046058458813109,
531
+ "loss": 0.9575,
532
+ "step": 820
533
+ },
534
+ {
535
+ "epoch": 1.06,
536
+ "learning_rate": 0.0002032772364924712,
537
+ "loss": 0.9826,
538
+ "step": 830
539
+ },
540
+ {
541
+ "epoch": 1.07,
542
+ "learning_rate": 0.0002019486271036315,
543
+ "loss": 0.973,
544
+ "step": 840
545
+ },
546
+ {
547
+ "epoch": 1.08,
548
+ "learning_rate": 0.00020062001771479183,
549
+ "loss": 0.9623,
550
+ "step": 850
551
+ },
552
+ {
553
+ "epoch": 1.09,
554
+ "learning_rate": 0.00019929140832595214,
555
+ "loss": 0.9618,
556
+ "step": 860
557
+ },
558
+ {
559
+ "epoch": 1.11,
560
+ "learning_rate": 0.00019796279893711247,
561
+ "loss": 0.9594,
562
+ "step": 870
563
+ },
564
+ {
565
+ "epoch": 1.12,
566
+ "learning_rate": 0.00019663418954827278,
567
+ "loss": 0.9827,
568
+ "step": 880
569
+ },
570
+ {
571
+ "epoch": 1.13,
572
+ "learning_rate": 0.0001953055801594331,
573
+ "loss": 0.9608,
574
+ "step": 890
575
+ },
576
+ {
577
+ "epoch": 1.14,
578
+ "learning_rate": 0.00019397697077059342,
579
+ "loss": 0.967,
580
+ "step": 900
581
+ },
582
+ {
583
+ "epoch": 1.16,
584
+ "learning_rate": 0.00019264836138175375,
585
+ "loss": 0.9498,
586
+ "step": 910
587
+ },
588
+ {
589
+ "epoch": 1.17,
590
+ "learning_rate": 0.00019131975199291405,
591
+ "loss": 0.972,
592
+ "step": 920
593
+ },
594
+ {
595
+ "epoch": 1.18,
596
+ "learning_rate": 0.0001899911426040744,
597
+ "loss": 0.9677,
598
+ "step": 930
599
+ },
600
+ {
601
+ "epoch": 1.2,
602
+ "learning_rate": 0.0001886625332152347,
603
+ "loss": 0.9715,
604
+ "step": 940
605
+ },
606
+ {
607
+ "epoch": 1.21,
608
+ "learning_rate": 0.00018733392382639503,
609
+ "loss": 0.9758,
610
+ "step": 950
611
+ },
612
+ {
613
+ "epoch": 1.22,
614
+ "learning_rate": 0.00018600531443755533,
615
+ "loss": 0.9667,
616
+ "step": 960
617
+ },
618
+ {
619
+ "epoch": 1.23,
620
+ "learning_rate": 0.00018467670504871564,
621
+ "loss": 0.9719,
622
+ "step": 970
623
+ },
624
+ {
625
+ "epoch": 1.25,
626
+ "learning_rate": 0.00018334809565987597,
627
+ "loss": 0.9584,
628
+ "step": 980
629
+ },
630
+ {
631
+ "epoch": 1.26,
632
+ "learning_rate": 0.00018201948627103628,
633
+ "loss": 0.9591,
634
+ "step": 990
635
+ },
636
+ {
637
+ "epoch": 1.27,
638
+ "learning_rate": 0.00018069087688219664,
639
+ "loss": 0.959,
640
+ "step": 1000
641
+ },
642
+ {
643
+ "epoch": 1.27,
644
+ "eval_loss": 0.9663413166999817,
645
+ "eval_runtime": 83.6093,
646
+ "eval_samples_per_second": 23.921,
647
+ "eval_steps_per_second": 0.754,
648
+ "step": 1000
649
+ },
650
+ {
651
+ "epoch": 1.28,
652
+ "learning_rate": 0.00017936226749335691,
653
+ "loss": 0.9545,
654
+ "step": 1010
655
+ },
656
+ {
657
+ "epoch": 1.3,
658
+ "learning_rate": 0.00017803365810451727,
659
+ "loss": 0.9675,
660
+ "step": 1020
661
+ },
662
+ {
663
+ "epoch": 1.31,
664
+ "learning_rate": 0.00017670504871567758,
665
+ "loss": 0.961,
666
+ "step": 1030
667
+ },
668
+ {
669
+ "epoch": 1.32,
670
+ "learning_rate": 0.0001753764393268379,
671
+ "loss": 0.9575,
672
+ "step": 1040
673
+ },
674
+ {
675
+ "epoch": 1.34,
676
+ "learning_rate": 0.00017404782993799822,
677
+ "loss": 0.9636,
678
+ "step": 1050
679
+ },
680
+ {
681
+ "epoch": 1.35,
682
+ "learning_rate": 0.00017271922054915855,
683
+ "loss": 0.9687,
684
+ "step": 1060
685
+ },
686
+ {
687
+ "epoch": 1.36,
688
+ "learning_rate": 0.00017139061116031886,
689
+ "loss": 0.9656,
690
+ "step": 1070
691
+ },
692
+ {
693
+ "epoch": 1.37,
694
+ "learning_rate": 0.0001700620017714792,
695
+ "loss": 0.9493,
696
+ "step": 1080
697
+ },
698
+ {
699
+ "epoch": 1.39,
700
+ "learning_rate": 0.0001687333923826395,
701
+ "loss": 0.9681,
702
+ "step": 1090
703
+ },
704
+ {
705
+ "epoch": 1.4,
706
+ "learning_rate": 0.00016740478299379983,
707
+ "loss": 0.9537,
708
+ "step": 1100
709
+ },
710
+ {
711
+ "epoch": 1.41,
712
+ "learning_rate": 0.00016607617360496013,
713
+ "loss": 0.9468,
714
+ "step": 1110
715
+ },
716
+ {
717
+ "epoch": 1.42,
718
+ "learning_rate": 0.00016474756421612044,
719
+ "loss": 0.957,
720
+ "step": 1120
721
+ },
722
+ {
723
+ "epoch": 1.44,
724
+ "learning_rate": 0.00016341895482728077,
725
+ "loss": 0.9697,
726
+ "step": 1130
727
+ },
728
+ {
729
+ "epoch": 1.45,
730
+ "learning_rate": 0.00016209034543844108,
731
+ "loss": 0.9689,
732
+ "step": 1140
733
+ },
734
+ {
735
+ "epoch": 1.46,
736
+ "learning_rate": 0.0001607617360496014,
737
+ "loss": 0.9691,
738
+ "step": 1150
739
+ },
740
+ {
741
+ "epoch": 1.47,
742
+ "learning_rate": 0.00015943312666076172,
743
+ "loss": 0.9678,
744
+ "step": 1160
745
+ },
746
+ {
747
+ "epoch": 1.49,
748
+ "learning_rate": 0.00015810451727192205,
749
+ "loss": 0.967,
750
+ "step": 1170
751
+ },
752
+ {
753
+ "epoch": 1.5,
754
+ "learning_rate": 0.00015677590788308235,
755
+ "loss": 0.9742,
756
+ "step": 1180
757
+ },
758
+ {
759
+ "epoch": 1.51,
760
+ "learning_rate": 0.0001554472984942427,
761
+ "loss": 0.9438,
762
+ "step": 1190
763
+ },
764
+ {
765
+ "epoch": 1.53,
766
+ "learning_rate": 0.000154118689105403,
767
+ "loss": 0.9688,
768
+ "step": 1200
769
+ },
770
+ {
771
+ "epoch": 1.53,
772
+ "eval_loss": 0.9622219204902649,
773
+ "eval_runtime": 78.9396,
774
+ "eval_samples_per_second": 25.336,
775
+ "eval_steps_per_second": 0.798,
776
+ "step": 1200
777
+ },
778
+ {
779
+ "epoch": 1.54,
780
+ "learning_rate": 0.00015279007971656333,
781
+ "loss": 0.9655,
782
+ "step": 1210
783
+ },
784
+ {
785
+ "epoch": 1.55,
786
+ "learning_rate": 0.00015146147032772363,
787
+ "loss": 0.957,
788
+ "step": 1220
789
+ },
790
+ {
791
+ "epoch": 1.56,
792
+ "learning_rate": 0.00015013286093888396,
793
+ "loss": 0.9667,
794
+ "step": 1230
795
+ },
796
+ {
797
+ "epoch": 1.58,
798
+ "learning_rate": 0.00014880425155004427,
799
+ "loss": 0.9554,
800
+ "step": 1240
801
+ },
802
+ {
803
+ "epoch": 1.59,
804
+ "learning_rate": 0.0001474756421612046,
805
+ "loss": 0.9691,
806
+ "step": 1250
807
+ },
808
+ {
809
+ "epoch": 1.6,
810
+ "learning_rate": 0.0001461470327723649,
811
+ "loss": 0.9587,
812
+ "step": 1260
813
+ },
814
+ {
815
+ "epoch": 1.61,
816
+ "learning_rate": 0.00014481842338352524,
817
+ "loss": 0.9461,
818
+ "step": 1270
819
+ },
820
+ {
821
+ "epoch": 1.63,
822
+ "learning_rate": 0.00014348981399468555,
823
+ "loss": 0.9607,
824
+ "step": 1280
825
+ },
826
+ {
827
+ "epoch": 1.64,
828
+ "learning_rate": 0.00014216120460584588,
829
+ "loss": 0.9528,
830
+ "step": 1290
831
+ },
832
+ {
833
+ "epoch": 1.65,
834
+ "learning_rate": 0.00014083259521700619,
835
+ "loss": 0.9469,
836
+ "step": 1300
837
+ },
838
+ {
839
+ "epoch": 1.67,
840
+ "learning_rate": 0.00013950398582816652,
841
+ "loss": 0.9415,
842
+ "step": 1310
843
+ },
844
+ {
845
+ "epoch": 1.68,
846
+ "learning_rate": 0.00013817537643932682,
847
+ "loss": 0.9628,
848
+ "step": 1320
849
+ },
850
+ {
851
+ "epoch": 1.69,
852
+ "learning_rate": 0.00013684676705048713,
853
+ "loss": 0.9562,
854
+ "step": 1330
855
+ },
856
+ {
857
+ "epoch": 1.7,
858
+ "learning_rate": 0.00013551815766164746,
859
+ "loss": 0.9579,
860
+ "step": 1340
861
+ },
862
+ {
863
+ "epoch": 1.72,
864
+ "learning_rate": 0.00013418954827280777,
865
+ "loss": 0.9578,
866
+ "step": 1350
867
+ },
868
+ {
869
+ "epoch": 1.73,
870
+ "learning_rate": 0.0001328609388839681,
871
+ "loss": 0.9508,
872
+ "step": 1360
873
+ },
874
+ {
875
+ "epoch": 1.74,
876
+ "learning_rate": 0.0001315323294951284,
877
+ "loss": 0.9596,
878
+ "step": 1370
879
+ },
880
+ {
881
+ "epoch": 1.75,
882
+ "learning_rate": 0.00013020372010628874,
883
+ "loss": 0.9446,
884
+ "step": 1380
885
+ },
886
+ {
887
+ "epoch": 1.77,
888
+ "learning_rate": 0.00012887511071744907,
889
+ "loss": 0.9418,
890
+ "step": 1390
891
+ },
892
+ {
893
+ "epoch": 1.78,
894
+ "learning_rate": 0.00012754650132860938,
895
+ "loss": 0.9447,
896
+ "step": 1400
897
+ },
898
+ {
899
+ "epoch": 1.78,
900
+ "eval_loss": 0.9596577286720276,
901
+ "eval_runtime": 78.8874,
902
+ "eval_samples_per_second": 25.353,
903
+ "eval_steps_per_second": 0.799,
904
+ "step": 1400
905
+ },
906
+ {
907
+ "epoch": 1.79,
908
+ "learning_rate": 0.0001262178919397697,
909
+ "loss": 0.9597,
910
+ "step": 1410
911
+ },
912
+ {
913
+ "epoch": 1.81,
914
+ "learning_rate": 0.00012488928255093002,
915
+ "loss": 0.9601,
916
+ "step": 1420
917
+ },
918
+ {
919
+ "epoch": 1.82,
920
+ "learning_rate": 0.00012356067316209035,
921
+ "loss": 0.9794,
922
+ "step": 1430
923
+ },
924
+ {
925
+ "epoch": 1.83,
926
+ "learning_rate": 0.00012223206377325066,
927
+ "loss": 0.9295,
928
+ "step": 1440
929
+ },
930
+ {
931
+ "epoch": 1.84,
932
+ "learning_rate": 0.00012090345438441097,
933
+ "loss": 0.9418,
934
+ "step": 1450
935
+ },
936
+ {
937
+ "epoch": 1.86,
938
+ "learning_rate": 0.00011957484499557128,
939
+ "loss": 0.941,
940
+ "step": 1460
941
+ },
942
+ {
943
+ "epoch": 1.87,
944
+ "learning_rate": 0.0001182462356067316,
945
+ "loss": 0.9504,
946
+ "step": 1470
947
+ },
948
+ {
949
+ "epoch": 1.88,
950
+ "learning_rate": 0.00011691762621789192,
951
+ "loss": 0.9489,
952
+ "step": 1480
953
+ },
954
+ {
955
+ "epoch": 1.89,
956
+ "learning_rate": 0.00011558901682905224,
957
+ "loss": 0.9482,
958
+ "step": 1490
959
+ },
960
+ {
961
+ "epoch": 1.91,
962
+ "learning_rate": 0.00011426040744021257,
963
+ "loss": 0.9435,
964
+ "step": 1500
965
+ },
966
+ {
967
+ "epoch": 1.92,
968
+ "learning_rate": 0.00011293179805137289,
969
+ "loss": 0.953,
970
+ "step": 1510
971
+ },
972
+ {
973
+ "epoch": 1.93,
974
+ "learning_rate": 0.00011160318866253321,
975
+ "loss": 0.9468,
976
+ "step": 1520
977
+ },
978
+ {
979
+ "epoch": 1.95,
980
+ "learning_rate": 0.00011027457927369353,
981
+ "loss": 0.9289,
982
+ "step": 1530
983
+ },
984
+ {
985
+ "epoch": 1.96,
986
+ "learning_rate": 0.00010894596988485385,
987
+ "loss": 0.9533,
988
+ "step": 1540
989
+ },
990
+ {
991
+ "epoch": 1.97,
992
+ "learning_rate": 0.00010761736049601417,
993
+ "loss": 0.9591,
994
+ "step": 1550
995
+ },
996
+ {
997
+ "epoch": 1.98,
998
+ "learning_rate": 0.00010628875110717449,
999
+ "loss": 0.951,
1000
+ "step": 1560
1001
+ },
1002
+ {
1003
+ "epoch": 2.0,
1004
+ "learning_rate": 0.0001049601417183348,
1005
+ "loss": 0.9171,
1006
+ "step": 1570
1007
+ },
1008
+ {
1009
+ "epoch": 2.01,
1010
+ "learning_rate": 0.00010363153232949512,
1011
+ "loss": 0.9375,
1012
+ "step": 1580
1013
+ },
1014
+ {
1015
+ "epoch": 2.02,
1016
+ "learning_rate": 0.00010230292294065544,
1017
+ "loss": 0.9568,
1018
+ "step": 1590
1019
+ },
1020
+ {
1021
+ "epoch": 2.03,
1022
+ "learning_rate": 0.00010097431355181575,
1023
+ "loss": 0.9633,
1024
+ "step": 1600
1025
+ },
1026
+ {
1027
+ "epoch": 2.03,
1028
+ "eval_loss": 0.9573116302490234,
1029
+ "eval_runtime": 79.0309,
1030
+ "eval_samples_per_second": 25.307,
1031
+ "eval_steps_per_second": 0.797,
1032
+ "step": 1600
1033
+ },
1034
+ {
1035
+ "epoch": 2.05,
1036
+ "learning_rate": 9.964570416297607e-05,
1037
+ "loss": 0.9424,
1038
+ "step": 1610
1039
+ },
1040
+ {
1041
+ "epoch": 2.06,
1042
+ "learning_rate": 9.831709477413639e-05,
1043
+ "loss": 0.9243,
1044
+ "step": 1620
1045
+ },
1046
+ {
1047
+ "epoch": 2.07,
1048
+ "learning_rate": 9.698848538529671e-05,
1049
+ "loss": 0.9415,
1050
+ "step": 1630
1051
+ },
1052
+ {
1053
+ "epoch": 2.09,
1054
+ "learning_rate": 9.565987599645703e-05,
1055
+ "loss": 0.9434,
1056
+ "step": 1640
1057
+ },
1058
+ {
1059
+ "epoch": 2.1,
1060
+ "learning_rate": 9.433126660761735e-05,
1061
+ "loss": 0.9467,
1062
+ "step": 1650
1063
+ },
1064
+ {
1065
+ "epoch": 2.11,
1066
+ "learning_rate": 9.300265721877767e-05,
1067
+ "loss": 0.9516,
1068
+ "step": 1660
1069
+ },
1070
+ {
1071
+ "epoch": 2.12,
1072
+ "learning_rate": 9.167404782993798e-05,
1073
+ "loss": 0.9617,
1074
+ "step": 1670
1075
+ },
1076
+ {
1077
+ "epoch": 2.14,
1078
+ "learning_rate": 9.034543844109832e-05,
1079
+ "loss": 0.9379,
1080
+ "step": 1680
1081
+ },
1082
+ {
1083
+ "epoch": 2.15,
1084
+ "learning_rate": 8.901682905225864e-05,
1085
+ "loss": 0.9455,
1086
+ "step": 1690
1087
+ },
1088
+ {
1089
+ "epoch": 2.16,
1090
+ "learning_rate": 8.768821966341896e-05,
1091
+ "loss": 0.9477,
1092
+ "step": 1700
1093
+ },
1094
+ {
1095
+ "epoch": 2.17,
1096
+ "learning_rate": 8.635961027457928e-05,
1097
+ "loss": 0.9448,
1098
+ "step": 1710
1099
+ },
1100
+ {
1101
+ "epoch": 2.19,
1102
+ "learning_rate": 8.50310008857396e-05,
1103
+ "loss": 0.9361,
1104
+ "step": 1720
1105
+ },
1106
+ {
1107
+ "epoch": 2.2,
1108
+ "learning_rate": 8.370239149689991e-05,
1109
+ "loss": 0.922,
1110
+ "step": 1730
1111
+ },
1112
+ {
1113
+ "epoch": 2.21,
1114
+ "learning_rate": 8.237378210806022e-05,
1115
+ "loss": 0.9442,
1116
+ "step": 1740
1117
+ },
1118
+ {
1119
+ "epoch": 2.23,
1120
+ "learning_rate": 8.104517271922054e-05,
1121
+ "loss": 0.9575,
1122
+ "step": 1750
1123
+ },
1124
+ {
1125
+ "epoch": 2.24,
1126
+ "learning_rate": 7.971656333038086e-05,
1127
+ "loss": 0.929,
1128
+ "step": 1760
1129
+ },
1130
+ {
1131
+ "epoch": 2.25,
1132
+ "learning_rate": 7.838795394154118e-05,
1133
+ "loss": 0.9312,
1134
+ "step": 1770
1135
+ },
1136
+ {
1137
+ "epoch": 2.26,
1138
+ "learning_rate": 7.70593445527015e-05,
1139
+ "loss": 0.943,
1140
+ "step": 1780
1141
+ },
1142
+ {
1143
+ "epoch": 2.28,
1144
+ "learning_rate": 7.573073516386182e-05,
1145
+ "loss": 0.9416,
1146
+ "step": 1790
1147
+ },
1148
+ {
1149
+ "epoch": 2.29,
1150
+ "learning_rate": 7.440212577502214e-05,
1151
+ "loss": 0.9455,
1152
+ "step": 1800
1153
+ },
1154
+ {
1155
+ "epoch": 2.29,
1156
+ "eval_loss": 0.9558648467063904,
1157
+ "eval_runtime": 79.0765,
1158
+ "eval_samples_per_second": 25.292,
1159
+ "eval_steps_per_second": 0.797,
1160
+ "step": 1800
1161
+ }
1162
+ ],
1163
+ "max_steps": 2358,
1164
+ "num_train_epochs": 3,
1165
+ "total_flos": 5.911772696235999e+18,
1166
+ "trial_name": null,
1167
+ "trial_params": null
1168
+ }
checkpoint-1800/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d466c45720cc9bfdf9011921a4cde00ee73d739fae921d7ae51485f2cfe2556d
3
+ size 3579
checkpoint-2000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e0e684264c8aa0a19130feadc93a27d0ccde69d15f40db8e46fa2f6bb9a01f8
3
+ size 104992197
checkpoint-2000/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ae5ea1222610a35288332588aa47075b81920eab1bac450c16e033508afff1ed
3
+ size 52486477
checkpoint-2000/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9414cbb31199100393f7f11fda6f1ca422b15759e4966be5416fa78ed73e99ac
3
+ size 14583
checkpoint-2000/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed6eb331f35b936a7216f933f9d4064c343b92e3bfba5703d86104f3995784e6
3
+ size 14583
checkpoint-2000/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5af0cf4ab7aa1016c826b6cd82278e5d547bece6a75c62b6d69551955a6312f2
3
+ size 14583
checkpoint-2000/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ea0a4cb4142c4528ccdd8b3de7c787ea9aae021713e824370fcdfb0d3804eeb
3
+ size 14583
checkpoint-2000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:02c7bd4df3e419c788455c072b5085a5a03ad060a38620f1ffac2cd1fec29f48
3
+ size 557
checkpoint-2000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42a6597101ec9035f4bbb93fada4c34dcda89e1b30ad129e51d7b3dd1a319ef3
3
+ size 627
checkpoint-2000/trainer_state.json ADDED
@@ -0,0 +1,1296 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9545624852180481,
3
+ "best_model_checkpoint": "./lora-alpaca/checkpoint-2000",
4
+ "epoch": 2.542911633820725,
5
+ "global_step": 2000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 2.6999999999999996e-05,
13
+ "loss": 1.6482,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.03,
18
+ "learning_rate": 5.399999999999999e-05,
19
+ "loss": 1.5884,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.04,
24
+ "learning_rate": 8.4e-05,
25
+ "loss": 1.5166,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.05,
30
+ "learning_rate": 0.00011399999999999999,
31
+ "loss": 1.3197,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.06,
36
+ "learning_rate": 0.00014399999999999998,
37
+ "loss": 1.1732,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.08,
42
+ "learning_rate": 0.00017399999999999997,
43
+ "loss": 1.1358,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.09,
48
+ "learning_rate": 0.000204,
49
+ "loss": 1.0792,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.1,
54
+ "learning_rate": 0.000234,
55
+ "loss": 1.062,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.11,
60
+ "learning_rate": 0.00026399999999999997,
61
+ "loss": 1.0665,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.13,
66
+ "learning_rate": 0.000294,
67
+ "loss": 1.0641,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.14,
72
+ "learning_rate": 0.00029893711248892826,
73
+ "loss": 1.0449,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.15,
78
+ "learning_rate": 0.00029760850310008854,
79
+ "loss": 1.0356,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.17,
84
+ "learning_rate": 0.0002962798937112489,
85
+ "loss": 1.0537,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.18,
90
+ "learning_rate": 0.0002949512843224092,
91
+ "loss": 1.0285,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.19,
96
+ "learning_rate": 0.0002936226749335695,
97
+ "loss": 1.0197,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.2,
102
+ "learning_rate": 0.0002922940655447298,
103
+ "loss": 1.0171,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.22,
108
+ "learning_rate": 0.00029096545615589015,
109
+ "loss": 1.0112,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.23,
114
+ "learning_rate": 0.0002896368467670505,
115
+ "loss": 1.0299,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.24,
120
+ "learning_rate": 0.00028830823737821076,
121
+ "loss": 1.0243,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.25,
126
+ "learning_rate": 0.0002869796279893711,
127
+ "loss": 1.0118,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.25,
132
+ "eval_loss": 1.0055800676345825,
133
+ "eval_runtime": 82.8279,
134
+ "eval_samples_per_second": 24.146,
135
+ "eval_steps_per_second": 0.761,
136
+ "step": 200
137
+ },
138
+ {
139
+ "epoch": 0.27,
140
+ "learning_rate": 0.00028565101860053143,
141
+ "loss": 0.9956,
142
+ "step": 210
143
+ },
144
+ {
145
+ "epoch": 0.28,
146
+ "learning_rate": 0.00028432240921169176,
147
+ "loss": 1.0039,
148
+ "step": 220
149
+ },
150
+ {
151
+ "epoch": 0.29,
152
+ "learning_rate": 0.00028299379982285204,
153
+ "loss": 1.0026,
154
+ "step": 230
155
+ },
156
+ {
157
+ "epoch": 0.31,
158
+ "learning_rate": 0.00028166519043401237,
159
+ "loss": 1.0148,
160
+ "step": 240
161
+ },
162
+ {
163
+ "epoch": 0.32,
164
+ "learning_rate": 0.0002803365810451727,
165
+ "loss": 1.0065,
166
+ "step": 250
167
+ },
168
+ {
169
+ "epoch": 0.33,
170
+ "learning_rate": 0.00027900797165633304,
171
+ "loss": 0.9965,
172
+ "step": 260
173
+ },
174
+ {
175
+ "epoch": 0.34,
176
+ "learning_rate": 0.0002776793622674933,
177
+ "loss": 1.0037,
178
+ "step": 270
179
+ },
180
+ {
181
+ "epoch": 0.36,
182
+ "learning_rate": 0.00027635075287865365,
183
+ "loss": 1.0149,
184
+ "step": 280
185
+ },
186
+ {
187
+ "epoch": 0.37,
188
+ "learning_rate": 0.000275022143489814,
189
+ "loss": 1.0136,
190
+ "step": 290
191
+ },
192
+ {
193
+ "epoch": 0.38,
194
+ "learning_rate": 0.00027369353410097426,
195
+ "loss": 0.9937,
196
+ "step": 300
197
+ },
198
+ {
199
+ "epoch": 0.39,
200
+ "learning_rate": 0.0002723649247121346,
201
+ "loss": 0.987,
202
+ "step": 310
203
+ },
204
+ {
205
+ "epoch": 0.41,
206
+ "learning_rate": 0.0002710363153232949,
207
+ "loss": 1.016,
208
+ "step": 320
209
+ },
210
+ {
211
+ "epoch": 0.42,
212
+ "learning_rate": 0.00026970770593445526,
213
+ "loss": 1.0136,
214
+ "step": 330
215
+ },
216
+ {
217
+ "epoch": 0.43,
218
+ "learning_rate": 0.00026837909654561554,
219
+ "loss": 0.9926,
220
+ "step": 340
221
+ },
222
+ {
223
+ "epoch": 0.45,
224
+ "learning_rate": 0.00026705048715677587,
225
+ "loss": 0.9922,
226
+ "step": 350
227
+ },
228
+ {
229
+ "epoch": 0.46,
230
+ "learning_rate": 0.0002657218777679362,
231
+ "loss": 0.9883,
232
+ "step": 360
233
+ },
234
+ {
235
+ "epoch": 0.47,
236
+ "learning_rate": 0.00026439326837909654,
237
+ "loss": 1.0076,
238
+ "step": 370
239
+ },
240
+ {
241
+ "epoch": 0.48,
242
+ "learning_rate": 0.0002630646589902568,
243
+ "loss": 0.9873,
244
+ "step": 380
245
+ },
246
+ {
247
+ "epoch": 0.5,
248
+ "learning_rate": 0.0002617360496014172,
249
+ "loss": 0.988,
250
+ "step": 390
251
+ },
252
+ {
253
+ "epoch": 0.51,
254
+ "learning_rate": 0.0002604074402125775,
255
+ "loss": 0.9902,
256
+ "step": 400
257
+ },
258
+ {
259
+ "epoch": 0.51,
260
+ "eval_loss": 0.9863091111183167,
261
+ "eval_runtime": 82.9481,
262
+ "eval_samples_per_second": 24.111,
263
+ "eval_steps_per_second": 0.76,
264
+ "step": 400
265
+ },
266
+ {
267
+ "epoch": 0.52,
268
+ "learning_rate": 0.00025907883082373776,
269
+ "loss": 0.9966,
270
+ "step": 410
271
+ },
272
+ {
273
+ "epoch": 0.53,
274
+ "learning_rate": 0.00025775022143489815,
275
+ "loss": 0.9914,
276
+ "step": 420
277
+ },
278
+ {
279
+ "epoch": 0.55,
280
+ "learning_rate": 0.0002564216120460584,
281
+ "loss": 0.9924,
282
+ "step": 430
283
+ },
284
+ {
285
+ "epoch": 0.56,
286
+ "learning_rate": 0.00025509300265721876,
287
+ "loss": 0.9948,
288
+ "step": 440
289
+ },
290
+ {
291
+ "epoch": 0.57,
292
+ "learning_rate": 0.0002537643932683791,
293
+ "loss": 0.9879,
294
+ "step": 450
295
+ },
296
+ {
297
+ "epoch": 0.58,
298
+ "learning_rate": 0.0002524357838795394,
299
+ "loss": 0.988,
300
+ "step": 460
301
+ },
302
+ {
303
+ "epoch": 0.6,
304
+ "learning_rate": 0.0002511071744906997,
305
+ "loss": 0.9928,
306
+ "step": 470
307
+ },
308
+ {
309
+ "epoch": 0.61,
310
+ "learning_rate": 0.00024977856510186003,
311
+ "loss": 0.9916,
312
+ "step": 480
313
+ },
314
+ {
315
+ "epoch": 0.62,
316
+ "learning_rate": 0.00024844995571302037,
317
+ "loss": 0.994,
318
+ "step": 490
319
+ },
320
+ {
321
+ "epoch": 0.64,
322
+ "learning_rate": 0.0002471213463241807,
323
+ "loss": 0.996,
324
+ "step": 500
325
+ },
326
+ {
327
+ "epoch": 0.65,
328
+ "learning_rate": 0.000245792736935341,
329
+ "loss": 0.9783,
330
+ "step": 510
331
+ },
332
+ {
333
+ "epoch": 0.66,
334
+ "learning_rate": 0.0002444641275465013,
335
+ "loss": 0.9798,
336
+ "step": 520
337
+ },
338
+ {
339
+ "epoch": 0.67,
340
+ "learning_rate": 0.00024313551815766162,
341
+ "loss": 0.9791,
342
+ "step": 530
343
+ },
344
+ {
345
+ "epoch": 0.69,
346
+ "learning_rate": 0.00024180690876882195,
347
+ "loss": 0.9894,
348
+ "step": 540
349
+ },
350
+ {
351
+ "epoch": 0.7,
352
+ "learning_rate": 0.00024047829937998225,
353
+ "loss": 0.9904,
354
+ "step": 550
355
+ },
356
+ {
357
+ "epoch": 0.71,
358
+ "learning_rate": 0.00023914968999114256,
359
+ "loss": 0.9942,
360
+ "step": 560
361
+ },
362
+ {
363
+ "epoch": 0.72,
364
+ "learning_rate": 0.0002378210806023029,
365
+ "loss": 0.9897,
366
+ "step": 570
367
+ },
368
+ {
369
+ "epoch": 0.74,
370
+ "learning_rate": 0.0002364924712134632,
371
+ "loss": 0.9741,
372
+ "step": 580
373
+ },
374
+ {
375
+ "epoch": 0.75,
376
+ "learning_rate": 0.00023516386182462353,
377
+ "loss": 0.9631,
378
+ "step": 590
379
+ },
380
+ {
381
+ "epoch": 0.76,
382
+ "learning_rate": 0.00023383525243578384,
383
+ "loss": 0.9784,
384
+ "step": 600
385
+ },
386
+ {
387
+ "epoch": 0.76,
388
+ "eval_loss": 0.9769166111946106,
389
+ "eval_runtime": 78.8243,
390
+ "eval_samples_per_second": 25.373,
391
+ "eval_steps_per_second": 0.799,
392
+ "step": 600
393
+ },
394
+ {
395
+ "epoch": 0.78,
396
+ "learning_rate": 0.0002325066430469442,
397
+ "loss": 0.9703,
398
+ "step": 610
399
+ },
400
+ {
401
+ "epoch": 0.79,
402
+ "learning_rate": 0.00023117803365810448,
403
+ "loss": 0.9914,
404
+ "step": 620
405
+ },
406
+ {
407
+ "epoch": 0.8,
408
+ "learning_rate": 0.00022984942426926484,
409
+ "loss": 0.9522,
410
+ "step": 630
411
+ },
412
+ {
413
+ "epoch": 0.81,
414
+ "learning_rate": 0.00022852081488042514,
415
+ "loss": 0.9773,
416
+ "step": 640
417
+ },
418
+ {
419
+ "epoch": 0.83,
420
+ "learning_rate": 0.00022719220549158547,
421
+ "loss": 0.9954,
422
+ "step": 650
423
+ },
424
+ {
425
+ "epoch": 0.84,
426
+ "learning_rate": 0.00022586359610274578,
427
+ "loss": 0.9768,
428
+ "step": 660
429
+ },
430
+ {
431
+ "epoch": 0.85,
432
+ "learning_rate": 0.0002245349867139061,
433
+ "loss": 0.9908,
434
+ "step": 670
435
+ },
436
+ {
437
+ "epoch": 0.86,
438
+ "learning_rate": 0.00022320637732506642,
439
+ "loss": 0.9944,
440
+ "step": 680
441
+ },
442
+ {
443
+ "epoch": 0.88,
444
+ "learning_rate": 0.00022187776793622672,
445
+ "loss": 0.9772,
446
+ "step": 690
447
+ },
448
+ {
449
+ "epoch": 0.89,
450
+ "learning_rate": 0.00022054915854738706,
451
+ "loss": 0.9697,
452
+ "step": 700
453
+ },
454
+ {
455
+ "epoch": 0.9,
456
+ "learning_rate": 0.00021922054915854736,
457
+ "loss": 0.9652,
458
+ "step": 710
459
+ },
460
+ {
461
+ "epoch": 0.92,
462
+ "learning_rate": 0.0002178919397697077,
463
+ "loss": 0.9738,
464
+ "step": 720
465
+ },
466
+ {
467
+ "epoch": 0.93,
468
+ "learning_rate": 0.000216563330380868,
469
+ "loss": 0.9892,
470
+ "step": 730
471
+ },
472
+ {
473
+ "epoch": 0.94,
474
+ "learning_rate": 0.00021523472099202833,
475
+ "loss": 0.9731,
476
+ "step": 740
477
+ },
478
+ {
479
+ "epoch": 0.95,
480
+ "learning_rate": 0.00021390611160318864,
481
+ "loss": 0.985,
482
+ "step": 750
483
+ },
484
+ {
485
+ "epoch": 0.97,
486
+ "learning_rate": 0.00021257750221434897,
487
+ "loss": 0.9918,
488
+ "step": 760
489
+ },
490
+ {
491
+ "epoch": 0.98,
492
+ "learning_rate": 0.00021124889282550928,
493
+ "loss": 0.9603,
494
+ "step": 770
495
+ },
496
+ {
497
+ "epoch": 0.99,
498
+ "learning_rate": 0.0002099202834366696,
499
+ "loss": 0.9686,
500
+ "step": 780
501
+ },
502
+ {
503
+ "epoch": 1.0,
504
+ "learning_rate": 0.00020859167404782992,
505
+ "loss": 0.9761,
506
+ "step": 790
507
+ },
508
+ {
509
+ "epoch": 1.02,
510
+ "learning_rate": 0.00020726306465899025,
511
+ "loss": 0.9812,
512
+ "step": 800
513
+ },
514
+ {
515
+ "epoch": 1.02,
516
+ "eval_loss": 0.970382571220398,
517
+ "eval_runtime": 78.9078,
518
+ "eval_samples_per_second": 25.346,
519
+ "eval_steps_per_second": 0.798,
520
+ "step": 800
521
+ },
522
+ {
523
+ "epoch": 1.03,
524
+ "learning_rate": 0.00020593445527015056,
525
+ "loss": 0.95,
526
+ "step": 810
527
+ },
528
+ {
529
+ "epoch": 1.04,
530
+ "learning_rate": 0.0002046058458813109,
531
+ "loss": 0.9575,
532
+ "step": 820
533
+ },
534
+ {
535
+ "epoch": 1.06,
536
+ "learning_rate": 0.0002032772364924712,
537
+ "loss": 0.9826,
538
+ "step": 830
539
+ },
540
+ {
541
+ "epoch": 1.07,
542
+ "learning_rate": 0.0002019486271036315,
543
+ "loss": 0.973,
544
+ "step": 840
545
+ },
546
+ {
547
+ "epoch": 1.08,
548
+ "learning_rate": 0.00020062001771479183,
549
+ "loss": 0.9623,
550
+ "step": 850
551
+ },
552
+ {
553
+ "epoch": 1.09,
554
+ "learning_rate": 0.00019929140832595214,
555
+ "loss": 0.9618,
556
+ "step": 860
557
+ },
558
+ {
559
+ "epoch": 1.11,
560
+ "learning_rate": 0.00019796279893711247,
561
+ "loss": 0.9594,
562
+ "step": 870
563
+ },
564
+ {
565
+ "epoch": 1.12,
566
+ "learning_rate": 0.00019663418954827278,
567
+ "loss": 0.9827,
568
+ "step": 880
569
+ },
570
+ {
571
+ "epoch": 1.13,
572
+ "learning_rate": 0.0001953055801594331,
573
+ "loss": 0.9608,
574
+ "step": 890
575
+ },
576
+ {
577
+ "epoch": 1.14,
578
+ "learning_rate": 0.00019397697077059342,
579
+ "loss": 0.967,
580
+ "step": 900
581
+ },
582
+ {
583
+ "epoch": 1.16,
584
+ "learning_rate": 0.00019264836138175375,
585
+ "loss": 0.9498,
586
+ "step": 910
587
+ },
588
+ {
589
+ "epoch": 1.17,
590
+ "learning_rate": 0.00019131975199291405,
591
+ "loss": 0.972,
592
+ "step": 920
593
+ },
594
+ {
595
+ "epoch": 1.18,
596
+ "learning_rate": 0.0001899911426040744,
597
+ "loss": 0.9677,
598
+ "step": 930
599
+ },
600
+ {
601
+ "epoch": 1.2,
602
+ "learning_rate": 0.0001886625332152347,
603
+ "loss": 0.9715,
604
+ "step": 940
605
+ },
606
+ {
607
+ "epoch": 1.21,
608
+ "learning_rate": 0.00018733392382639503,
609
+ "loss": 0.9758,
610
+ "step": 950
611
+ },
612
+ {
613
+ "epoch": 1.22,
614
+ "learning_rate": 0.00018600531443755533,
615
+ "loss": 0.9667,
616
+ "step": 960
617
+ },
618
+ {
619
+ "epoch": 1.23,
620
+ "learning_rate": 0.00018467670504871564,
621
+ "loss": 0.9719,
622
+ "step": 970
623
+ },
624
+ {
625
+ "epoch": 1.25,
626
+ "learning_rate": 0.00018334809565987597,
627
+ "loss": 0.9584,
628
+ "step": 980
629
+ },
630
+ {
631
+ "epoch": 1.26,
632
+ "learning_rate": 0.00018201948627103628,
633
+ "loss": 0.9591,
634
+ "step": 990
635
+ },
636
+ {
637
+ "epoch": 1.27,
638
+ "learning_rate": 0.00018069087688219664,
639
+ "loss": 0.959,
640
+ "step": 1000
641
+ },
642
+ {
643
+ "epoch": 1.27,
644
+ "eval_loss": 0.9663413166999817,
645
+ "eval_runtime": 83.6093,
646
+ "eval_samples_per_second": 23.921,
647
+ "eval_steps_per_second": 0.754,
648
+ "step": 1000
649
+ },
650
+ {
651
+ "epoch": 1.28,
652
+ "learning_rate": 0.00017936226749335691,
653
+ "loss": 0.9545,
654
+ "step": 1010
655
+ },
656
+ {
657
+ "epoch": 1.3,
658
+ "learning_rate": 0.00017803365810451727,
659
+ "loss": 0.9675,
660
+ "step": 1020
661
+ },
662
+ {
663
+ "epoch": 1.31,
664
+ "learning_rate": 0.00017670504871567758,
665
+ "loss": 0.961,
666
+ "step": 1030
667
+ },
668
+ {
669
+ "epoch": 1.32,
670
+ "learning_rate": 0.0001753764393268379,
671
+ "loss": 0.9575,
672
+ "step": 1040
673
+ },
674
+ {
675
+ "epoch": 1.34,
676
+ "learning_rate": 0.00017404782993799822,
677
+ "loss": 0.9636,
678
+ "step": 1050
679
+ },
680
+ {
681
+ "epoch": 1.35,
682
+ "learning_rate": 0.00017271922054915855,
683
+ "loss": 0.9687,
684
+ "step": 1060
685
+ },
686
+ {
687
+ "epoch": 1.36,
688
+ "learning_rate": 0.00017139061116031886,
689
+ "loss": 0.9656,
690
+ "step": 1070
691
+ },
692
+ {
693
+ "epoch": 1.37,
694
+ "learning_rate": 0.0001700620017714792,
695
+ "loss": 0.9493,
696
+ "step": 1080
697
+ },
698
+ {
699
+ "epoch": 1.39,
700
+ "learning_rate": 0.0001687333923826395,
701
+ "loss": 0.9681,
702
+ "step": 1090
703
+ },
704
+ {
705
+ "epoch": 1.4,
706
+ "learning_rate": 0.00016740478299379983,
707
+ "loss": 0.9537,
708
+ "step": 1100
709
+ },
710
+ {
711
+ "epoch": 1.41,
712
+ "learning_rate": 0.00016607617360496013,
713
+ "loss": 0.9468,
714
+ "step": 1110
715
+ },
716
+ {
717
+ "epoch": 1.42,
718
+ "learning_rate": 0.00016474756421612044,
719
+ "loss": 0.957,
720
+ "step": 1120
721
+ },
722
+ {
723
+ "epoch": 1.44,
724
+ "learning_rate": 0.00016341895482728077,
725
+ "loss": 0.9697,
726
+ "step": 1130
727
+ },
728
+ {
729
+ "epoch": 1.45,
730
+ "learning_rate": 0.00016209034543844108,
731
+ "loss": 0.9689,
732
+ "step": 1140
733
+ },
734
+ {
735
+ "epoch": 1.46,
736
+ "learning_rate": 0.0001607617360496014,
737
+ "loss": 0.9691,
738
+ "step": 1150
739
+ },
740
+ {
741
+ "epoch": 1.47,
742
+ "learning_rate": 0.00015943312666076172,
743
+ "loss": 0.9678,
744
+ "step": 1160
745
+ },
746
+ {
747
+ "epoch": 1.49,
748
+ "learning_rate": 0.00015810451727192205,
749
+ "loss": 0.967,
750
+ "step": 1170
751
+ },
752
+ {
753
+ "epoch": 1.5,
754
+ "learning_rate": 0.00015677590788308235,
755
+ "loss": 0.9742,
756
+ "step": 1180
757
+ },
758
+ {
759
+ "epoch": 1.51,
760
+ "learning_rate": 0.0001554472984942427,
761
+ "loss": 0.9438,
762
+ "step": 1190
763
+ },
764
+ {
765
+ "epoch": 1.53,
766
+ "learning_rate": 0.000154118689105403,
767
+ "loss": 0.9688,
768
+ "step": 1200
769
+ },
770
+ {
771
+ "epoch": 1.53,
772
+ "eval_loss": 0.9622219204902649,
773
+ "eval_runtime": 78.9396,
774
+ "eval_samples_per_second": 25.336,
775
+ "eval_steps_per_second": 0.798,
776
+ "step": 1200
777
+ },
778
+ {
779
+ "epoch": 1.54,
780
+ "learning_rate": 0.00015279007971656333,
781
+ "loss": 0.9655,
782
+ "step": 1210
783
+ },
784
+ {
785
+ "epoch": 1.55,
786
+ "learning_rate": 0.00015146147032772363,
787
+ "loss": 0.957,
788
+ "step": 1220
789
+ },
790
+ {
791
+ "epoch": 1.56,
792
+ "learning_rate": 0.00015013286093888396,
793
+ "loss": 0.9667,
794
+ "step": 1230
795
+ },
796
+ {
797
+ "epoch": 1.58,
798
+ "learning_rate": 0.00014880425155004427,
799
+ "loss": 0.9554,
800
+ "step": 1240
801
+ },
802
+ {
803
+ "epoch": 1.59,
804
+ "learning_rate": 0.0001474756421612046,
805
+ "loss": 0.9691,
806
+ "step": 1250
807
+ },
808
+ {
809
+ "epoch": 1.6,
810
+ "learning_rate": 0.0001461470327723649,
811
+ "loss": 0.9587,
812
+ "step": 1260
813
+ },
814
+ {
815
+ "epoch": 1.61,
816
+ "learning_rate": 0.00014481842338352524,
817
+ "loss": 0.9461,
818
+ "step": 1270
819
+ },
820
+ {
821
+ "epoch": 1.63,
822
+ "learning_rate": 0.00014348981399468555,
823
+ "loss": 0.9607,
824
+ "step": 1280
825
+ },
826
+ {
827
+ "epoch": 1.64,
828
+ "learning_rate": 0.00014216120460584588,
829
+ "loss": 0.9528,
830
+ "step": 1290
831
+ },
832
+ {
833
+ "epoch": 1.65,
834
+ "learning_rate": 0.00014083259521700619,
835
+ "loss": 0.9469,
836
+ "step": 1300
837
+ },
838
+ {
839
+ "epoch": 1.67,
840
+ "learning_rate": 0.00013950398582816652,
841
+ "loss": 0.9415,
842
+ "step": 1310
843
+ },
844
+ {
845
+ "epoch": 1.68,
846
+ "learning_rate": 0.00013817537643932682,
847
+ "loss": 0.9628,
848
+ "step": 1320
849
+ },
850
+ {
851
+ "epoch": 1.69,
852
+ "learning_rate": 0.00013684676705048713,
853
+ "loss": 0.9562,
854
+ "step": 1330
855
+ },
856
+ {
857
+ "epoch": 1.7,
858
+ "learning_rate": 0.00013551815766164746,
859
+ "loss": 0.9579,
860
+ "step": 1340
861
+ },
862
+ {
863
+ "epoch": 1.72,
864
+ "learning_rate": 0.00013418954827280777,
865
+ "loss": 0.9578,
866
+ "step": 1350
867
+ },
868
+ {
869
+ "epoch": 1.73,
870
+ "learning_rate": 0.0001328609388839681,
871
+ "loss": 0.9508,
872
+ "step": 1360
873
+ },
874
+ {
875
+ "epoch": 1.74,
876
+ "learning_rate": 0.0001315323294951284,
877
+ "loss": 0.9596,
878
+ "step": 1370
879
+ },
880
+ {
881
+ "epoch": 1.75,
882
+ "learning_rate": 0.00013020372010628874,
883
+ "loss": 0.9446,
884
+ "step": 1380
885
+ },
886
+ {
887
+ "epoch": 1.77,
888
+ "learning_rate": 0.00012887511071744907,
889
+ "loss": 0.9418,
890
+ "step": 1390
891
+ },
892
+ {
893
+ "epoch": 1.78,
894
+ "learning_rate": 0.00012754650132860938,
895
+ "loss": 0.9447,
896
+ "step": 1400
897
+ },
898
+ {
899
+ "epoch": 1.78,
900
+ "eval_loss": 0.9596577286720276,
901
+ "eval_runtime": 78.8874,
902
+ "eval_samples_per_second": 25.353,
903
+ "eval_steps_per_second": 0.799,
904
+ "step": 1400
905
+ },
906
+ {
907
+ "epoch": 1.79,
908
+ "learning_rate": 0.0001262178919397697,
909
+ "loss": 0.9597,
910
+ "step": 1410
911
+ },
912
+ {
913
+ "epoch": 1.81,
914
+ "learning_rate": 0.00012488928255093002,
915
+ "loss": 0.9601,
916
+ "step": 1420
917
+ },
918
+ {
919
+ "epoch": 1.82,
920
+ "learning_rate": 0.00012356067316209035,
921
+ "loss": 0.9794,
922
+ "step": 1430
923
+ },
924
+ {
925
+ "epoch": 1.83,
926
+ "learning_rate": 0.00012223206377325066,
927
+ "loss": 0.9295,
928
+ "step": 1440
929
+ },
930
+ {
931
+ "epoch": 1.84,
932
+ "learning_rate": 0.00012090345438441097,
933
+ "loss": 0.9418,
934
+ "step": 1450
935
+ },
936
+ {
937
+ "epoch": 1.86,
938
+ "learning_rate": 0.00011957484499557128,
939
+ "loss": 0.941,
940
+ "step": 1460
941
+ },
942
+ {
943
+ "epoch": 1.87,
944
+ "learning_rate": 0.0001182462356067316,
945
+ "loss": 0.9504,
946
+ "step": 1470
947
+ },
948
+ {
949
+ "epoch": 1.88,
950
+ "learning_rate": 0.00011691762621789192,
951
+ "loss": 0.9489,
952
+ "step": 1480
953
+ },
954
+ {
955
+ "epoch": 1.89,
956
+ "learning_rate": 0.00011558901682905224,
957
+ "loss": 0.9482,
958
+ "step": 1490
959
+ },
960
+ {
961
+ "epoch": 1.91,
962
+ "learning_rate": 0.00011426040744021257,
963
+ "loss": 0.9435,
964
+ "step": 1500
965
+ },
966
+ {
967
+ "epoch": 1.92,
968
+ "learning_rate": 0.00011293179805137289,
969
+ "loss": 0.953,
970
+ "step": 1510
971
+ },
972
+ {
973
+ "epoch": 1.93,
974
+ "learning_rate": 0.00011160318866253321,
975
+ "loss": 0.9468,
976
+ "step": 1520
977
+ },
978
+ {
979
+ "epoch": 1.95,
980
+ "learning_rate": 0.00011027457927369353,
981
+ "loss": 0.9289,
982
+ "step": 1530
983
+ },
984
+ {
985
+ "epoch": 1.96,
986
+ "learning_rate": 0.00010894596988485385,
987
+ "loss": 0.9533,
988
+ "step": 1540
989
+ },
990
+ {
991
+ "epoch": 1.97,
992
+ "learning_rate": 0.00010761736049601417,
993
+ "loss": 0.9591,
994
+ "step": 1550
995
+ },
996
+ {
997
+ "epoch": 1.98,
998
+ "learning_rate": 0.00010628875110717449,
999
+ "loss": 0.951,
1000
+ "step": 1560
1001
+ },
1002
+ {
1003
+ "epoch": 2.0,
1004
+ "learning_rate": 0.0001049601417183348,
1005
+ "loss": 0.9171,
1006
+ "step": 1570
1007
+ },
1008
+ {
1009
+ "epoch": 2.01,
1010
+ "learning_rate": 0.00010363153232949512,
1011
+ "loss": 0.9375,
1012
+ "step": 1580
1013
+ },
1014
+ {
1015
+ "epoch": 2.02,
1016
+ "learning_rate": 0.00010230292294065544,
1017
+ "loss": 0.9568,
1018
+ "step": 1590
1019
+ },
1020
+ {
1021
+ "epoch": 2.03,
1022
+ "learning_rate": 0.00010097431355181575,
1023
+ "loss": 0.9633,
1024
+ "step": 1600
1025
+ },
1026
+ {
1027
+ "epoch": 2.03,
1028
+ "eval_loss": 0.9573116302490234,
1029
+ "eval_runtime": 79.0309,
1030
+ "eval_samples_per_second": 25.307,
1031
+ "eval_steps_per_second": 0.797,
1032
+ "step": 1600
1033
+ },
1034
+ {
1035
+ "epoch": 2.05,
1036
+ "learning_rate": 9.964570416297607e-05,
1037
+ "loss": 0.9424,
1038
+ "step": 1610
1039
+ },
1040
+ {
1041
+ "epoch": 2.06,
1042
+ "learning_rate": 9.831709477413639e-05,
1043
+ "loss": 0.9243,
1044
+ "step": 1620
1045
+ },
1046
+ {
1047
+ "epoch": 2.07,
1048
+ "learning_rate": 9.698848538529671e-05,
1049
+ "loss": 0.9415,
1050
+ "step": 1630
1051
+ },
1052
+ {
1053
+ "epoch": 2.09,
1054
+ "learning_rate": 9.565987599645703e-05,
1055
+ "loss": 0.9434,
1056
+ "step": 1640
1057
+ },
1058
+ {
1059
+ "epoch": 2.1,
1060
+ "learning_rate": 9.433126660761735e-05,
1061
+ "loss": 0.9467,
1062
+ "step": 1650
1063
+ },
1064
+ {
1065
+ "epoch": 2.11,
1066
+ "learning_rate": 9.300265721877767e-05,
1067
+ "loss": 0.9516,
1068
+ "step": 1660
1069
+ },
1070
+ {
1071
+ "epoch": 2.12,
1072
+ "learning_rate": 9.167404782993798e-05,
1073
+ "loss": 0.9617,
1074
+ "step": 1670
1075
+ },
1076
+ {
1077
+ "epoch": 2.14,
1078
+ "learning_rate": 9.034543844109832e-05,
1079
+ "loss": 0.9379,
1080
+ "step": 1680
1081
+ },
1082
+ {
1083
+ "epoch": 2.15,
1084
+ "learning_rate": 8.901682905225864e-05,
1085
+ "loss": 0.9455,
1086
+ "step": 1690
1087
+ },
1088
+ {
1089
+ "epoch": 2.16,
1090
+ "learning_rate": 8.768821966341896e-05,
1091
+ "loss": 0.9477,
1092
+ "step": 1700
1093
+ },
1094
+ {
1095
+ "epoch": 2.17,
1096
+ "learning_rate": 8.635961027457928e-05,
1097
+ "loss": 0.9448,
1098
+ "step": 1710
1099
+ },
1100
+ {
1101
+ "epoch": 2.19,
1102
+ "learning_rate": 8.50310008857396e-05,
1103
+ "loss": 0.9361,
1104
+ "step": 1720
1105
+ },
1106
+ {
1107
+ "epoch": 2.2,
1108
+ "learning_rate": 8.370239149689991e-05,
1109
+ "loss": 0.922,
1110
+ "step": 1730
1111
+ },
1112
+ {
1113
+ "epoch": 2.21,
1114
+ "learning_rate": 8.237378210806022e-05,
1115
+ "loss": 0.9442,
1116
+ "step": 1740
1117
+ },
1118
+ {
1119
+ "epoch": 2.23,
1120
+ "learning_rate": 8.104517271922054e-05,
1121
+ "loss": 0.9575,
1122
+ "step": 1750
1123
+ },
1124
+ {
1125
+ "epoch": 2.24,
1126
+ "learning_rate": 7.971656333038086e-05,
1127
+ "loss": 0.929,
1128
+ "step": 1760
1129
+ },
1130
+ {
1131
+ "epoch": 2.25,
1132
+ "learning_rate": 7.838795394154118e-05,
1133
+ "loss": 0.9312,
1134
+ "step": 1770
1135
+ },
1136
+ {
1137
+ "epoch": 2.26,
1138
+ "learning_rate": 7.70593445527015e-05,
1139
+ "loss": 0.943,
1140
+ "step": 1780
1141
+ },
1142
+ {
1143
+ "epoch": 2.28,
1144
+ "learning_rate": 7.573073516386182e-05,
1145
+ "loss": 0.9416,
1146
+ "step": 1790
1147
+ },
1148
+ {
1149
+ "epoch": 2.29,
1150
+ "learning_rate": 7.440212577502214e-05,
1151
+ "loss": 0.9455,
1152
+ "step": 1800
1153
+ },
1154
+ {
1155
+ "epoch": 2.29,
1156
+ "eval_loss": 0.9558648467063904,
1157
+ "eval_runtime": 79.0765,
1158
+ "eval_samples_per_second": 25.292,
1159
+ "eval_steps_per_second": 0.797,
1160
+ "step": 1800
1161
+ },
1162
+ {
1163
+ "epoch": 2.3,
1164
+ "learning_rate": 7.307351638618245e-05,
1165
+ "loss": 0.9307,
1166
+ "step": 1810
1167
+ },
1168
+ {
1169
+ "epoch": 2.31,
1170
+ "learning_rate": 7.174490699734277e-05,
1171
+ "loss": 0.9517,
1172
+ "step": 1820
1173
+ },
1174
+ {
1175
+ "epoch": 2.33,
1176
+ "learning_rate": 7.041629760850309e-05,
1177
+ "loss": 0.9298,
1178
+ "step": 1830
1179
+ },
1180
+ {
1181
+ "epoch": 2.34,
1182
+ "learning_rate": 6.908768821966341e-05,
1183
+ "loss": 0.936,
1184
+ "step": 1840
1185
+ },
1186
+ {
1187
+ "epoch": 2.35,
1188
+ "learning_rate": 6.775907883082373e-05,
1189
+ "loss": 0.944,
1190
+ "step": 1850
1191
+ },
1192
+ {
1193
+ "epoch": 2.36,
1194
+ "learning_rate": 6.643046944198405e-05,
1195
+ "loss": 0.9521,
1196
+ "step": 1860
1197
+ },
1198
+ {
1199
+ "epoch": 2.38,
1200
+ "learning_rate": 6.510186005314437e-05,
1201
+ "loss": 0.9579,
1202
+ "step": 1870
1203
+ },
1204
+ {
1205
+ "epoch": 2.39,
1206
+ "learning_rate": 6.377325066430469e-05,
1207
+ "loss": 0.9412,
1208
+ "step": 1880
1209
+ },
1210
+ {
1211
+ "epoch": 2.4,
1212
+ "learning_rate": 6.244464127546501e-05,
1213
+ "loss": 0.9268,
1214
+ "step": 1890
1215
+ },
1216
+ {
1217
+ "epoch": 2.42,
1218
+ "learning_rate": 6.111603188662533e-05,
1219
+ "loss": 0.9343,
1220
+ "step": 1900
1221
+ },
1222
+ {
1223
+ "epoch": 2.43,
1224
+ "learning_rate": 5.978742249778564e-05,
1225
+ "loss": 0.9517,
1226
+ "step": 1910
1227
+ },
1228
+ {
1229
+ "epoch": 2.44,
1230
+ "learning_rate": 5.845881310894596e-05,
1231
+ "loss": 0.944,
1232
+ "step": 1920
1233
+ },
1234
+ {
1235
+ "epoch": 2.45,
1236
+ "learning_rate": 5.7130203720106285e-05,
1237
+ "loss": 0.9384,
1238
+ "step": 1930
1239
+ },
1240
+ {
1241
+ "epoch": 2.47,
1242
+ "learning_rate": 5.5801594331266605e-05,
1243
+ "loss": 0.9462,
1244
+ "step": 1940
1245
+ },
1246
+ {
1247
+ "epoch": 2.48,
1248
+ "learning_rate": 5.4472984942426924e-05,
1249
+ "loss": 0.9317,
1250
+ "step": 1950
1251
+ },
1252
+ {
1253
+ "epoch": 2.49,
1254
+ "learning_rate": 5.314437555358724e-05,
1255
+ "loss": 0.9354,
1256
+ "step": 1960
1257
+ },
1258
+ {
1259
+ "epoch": 2.5,
1260
+ "learning_rate": 5.181576616474756e-05,
1261
+ "loss": 0.9261,
1262
+ "step": 1970
1263
+ },
1264
+ {
1265
+ "epoch": 2.52,
1266
+ "learning_rate": 5.0487156775907875e-05,
1267
+ "loss": 0.9317,
1268
+ "step": 1980
1269
+ },
1270
+ {
1271
+ "epoch": 2.53,
1272
+ "learning_rate": 4.9158547387068194e-05,
1273
+ "loss": 0.932,
1274
+ "step": 1990
1275
+ },
1276
+ {
1277
+ "epoch": 2.54,
1278
+ "learning_rate": 4.7829937998228513e-05,
1279
+ "loss": 0.9323,
1280
+ "step": 2000
1281
+ },
1282
+ {
1283
+ "epoch": 2.54,
1284
+ "eval_loss": 0.9545624852180481,
1285
+ "eval_runtime": 78.9727,
1286
+ "eval_samples_per_second": 25.325,
1287
+ "eval_steps_per_second": 0.798,
1288
+ "step": 2000
1289
+ }
1290
+ ],
1291
+ "max_steps": 2358,
1292
+ "num_train_epochs": 3,
1293
+ "total_flos": 6.572646890393829e+18,
1294
+ "trial_name": null,
1295
+ "trial_params": null
1296
+ }
checkpoint-2000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d466c45720cc9bfdf9011921a4cde00ee73d739fae921d7ae51485f2cfe2556d
3
+ size 3579
checkpoint-2200/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e57ff6dbb35645df940be3b733c95dc577824978c2751e07e230fa004aabc0b4
3
+ size 104992197
checkpoint-2200/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9084d47a0042d54033081576d2c0ab83f27277ce67e2031dd989cf028180724e
3
+ size 52486477
checkpoint-2200/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:57d7104c582090da744316a40bce057f383c5b5facb057851efa46859fe125a1
3
+ size 14583
checkpoint-2200/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4de1f128f489d71f932dc381792ea034e8e8bf174350a93f2b8020612f715be9
3
+ size 14583
checkpoint-2200/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e3b68646f3a6fe0282559c54f21eab68813f26b26be63bfdf0c8b61d8558f245
3
+ size 14583
checkpoint-2200/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6f073e4fcf5e94598eaa3f83f5fc2342d8cba50a3ff6385aa5239ef2397bea0c
3
+ size 14583
checkpoint-2200/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c58ca96d689aea01ae4aab6b1ba8a77a96c9459bb1b7b3d6a093312999e27aa8
3
+ size 557
checkpoint-2200/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d11b4a104ba15216f3d021df8290946e26bb791f625d5527ada4e46123877668
3
+ size 627
checkpoint-2200/trainer_state.json ADDED
@@ -0,0 +1,1424 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.953123152256012,
3
+ "best_model_checkpoint": "./lora-alpaca/checkpoint-2200",
4
+ "epoch": 2.797202797202797,
5
+ "global_step": 2200,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 2.6999999999999996e-05,
13
+ "loss": 1.6482,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.03,
18
+ "learning_rate": 5.399999999999999e-05,
19
+ "loss": 1.5884,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.04,
24
+ "learning_rate": 8.4e-05,
25
+ "loss": 1.5166,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.05,
30
+ "learning_rate": 0.00011399999999999999,
31
+ "loss": 1.3197,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.06,
36
+ "learning_rate": 0.00014399999999999998,
37
+ "loss": 1.1732,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.08,
42
+ "learning_rate": 0.00017399999999999997,
43
+ "loss": 1.1358,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.09,
48
+ "learning_rate": 0.000204,
49
+ "loss": 1.0792,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.1,
54
+ "learning_rate": 0.000234,
55
+ "loss": 1.062,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.11,
60
+ "learning_rate": 0.00026399999999999997,
61
+ "loss": 1.0665,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.13,
66
+ "learning_rate": 0.000294,
67
+ "loss": 1.0641,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.14,
72
+ "learning_rate": 0.00029893711248892826,
73
+ "loss": 1.0449,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.15,
78
+ "learning_rate": 0.00029760850310008854,
79
+ "loss": 1.0356,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.17,
84
+ "learning_rate": 0.0002962798937112489,
85
+ "loss": 1.0537,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.18,
90
+ "learning_rate": 0.0002949512843224092,
91
+ "loss": 1.0285,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.19,
96
+ "learning_rate": 0.0002936226749335695,
97
+ "loss": 1.0197,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.2,
102
+ "learning_rate": 0.0002922940655447298,
103
+ "loss": 1.0171,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.22,
108
+ "learning_rate": 0.00029096545615589015,
109
+ "loss": 1.0112,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.23,
114
+ "learning_rate": 0.0002896368467670505,
115
+ "loss": 1.0299,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.24,
120
+ "learning_rate": 0.00028830823737821076,
121
+ "loss": 1.0243,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.25,
126
+ "learning_rate": 0.0002869796279893711,
127
+ "loss": 1.0118,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.25,
132
+ "eval_loss": 1.0055800676345825,
133
+ "eval_runtime": 82.8279,
134
+ "eval_samples_per_second": 24.146,
135
+ "eval_steps_per_second": 0.761,
136
+ "step": 200
137
+ },
138
+ {
139
+ "epoch": 0.27,
140
+ "learning_rate": 0.00028565101860053143,
141
+ "loss": 0.9956,
142
+ "step": 210
143
+ },
144
+ {
145
+ "epoch": 0.28,
146
+ "learning_rate": 0.00028432240921169176,
147
+ "loss": 1.0039,
148
+ "step": 220
149
+ },
150
+ {
151
+ "epoch": 0.29,
152
+ "learning_rate": 0.00028299379982285204,
153
+ "loss": 1.0026,
154
+ "step": 230
155
+ },
156
+ {
157
+ "epoch": 0.31,
158
+ "learning_rate": 0.00028166519043401237,
159
+ "loss": 1.0148,
160
+ "step": 240
161
+ },
162
+ {
163
+ "epoch": 0.32,
164
+ "learning_rate": 0.0002803365810451727,
165
+ "loss": 1.0065,
166
+ "step": 250
167
+ },
168
+ {
169
+ "epoch": 0.33,
170
+ "learning_rate": 0.00027900797165633304,
171
+ "loss": 0.9965,
172
+ "step": 260
173
+ },
174
+ {
175
+ "epoch": 0.34,
176
+ "learning_rate": 0.0002776793622674933,
177
+ "loss": 1.0037,
178
+ "step": 270
179
+ },
180
+ {
181
+ "epoch": 0.36,
182
+ "learning_rate": 0.00027635075287865365,
183
+ "loss": 1.0149,
184
+ "step": 280
185
+ },
186
+ {
187
+ "epoch": 0.37,
188
+ "learning_rate": 0.000275022143489814,
189
+ "loss": 1.0136,
190
+ "step": 290
191
+ },
192
+ {
193
+ "epoch": 0.38,
194
+ "learning_rate": 0.00027369353410097426,
195
+ "loss": 0.9937,
196
+ "step": 300
197
+ },
198
+ {
199
+ "epoch": 0.39,
200
+ "learning_rate": 0.0002723649247121346,
201
+ "loss": 0.987,
202
+ "step": 310
203
+ },
204
+ {
205
+ "epoch": 0.41,
206
+ "learning_rate": 0.0002710363153232949,
207
+ "loss": 1.016,
208
+ "step": 320
209
+ },
210
+ {
211
+ "epoch": 0.42,
212
+ "learning_rate": 0.00026970770593445526,
213
+ "loss": 1.0136,
214
+ "step": 330
215
+ },
216
+ {
217
+ "epoch": 0.43,
218
+ "learning_rate": 0.00026837909654561554,
219
+ "loss": 0.9926,
220
+ "step": 340
221
+ },
222
+ {
223
+ "epoch": 0.45,
224
+ "learning_rate": 0.00026705048715677587,
225
+ "loss": 0.9922,
226
+ "step": 350
227
+ },
228
+ {
229
+ "epoch": 0.46,
230
+ "learning_rate": 0.0002657218777679362,
231
+ "loss": 0.9883,
232
+ "step": 360
233
+ },
234
+ {
235
+ "epoch": 0.47,
236
+ "learning_rate": 0.00026439326837909654,
237
+ "loss": 1.0076,
238
+ "step": 370
239
+ },
240
+ {
241
+ "epoch": 0.48,
242
+ "learning_rate": 0.0002630646589902568,
243
+ "loss": 0.9873,
244
+ "step": 380
245
+ },
246
+ {
247
+ "epoch": 0.5,
248
+ "learning_rate": 0.0002617360496014172,
249
+ "loss": 0.988,
250
+ "step": 390
251
+ },
252
+ {
253
+ "epoch": 0.51,
254
+ "learning_rate": 0.0002604074402125775,
255
+ "loss": 0.9902,
256
+ "step": 400
257
+ },
258
+ {
259
+ "epoch": 0.51,
260
+ "eval_loss": 0.9863091111183167,
261
+ "eval_runtime": 82.9481,
262
+ "eval_samples_per_second": 24.111,
263
+ "eval_steps_per_second": 0.76,
264
+ "step": 400
265
+ },
266
+ {
267
+ "epoch": 0.52,
268
+ "learning_rate": 0.00025907883082373776,
269
+ "loss": 0.9966,
270
+ "step": 410
271
+ },
272
+ {
273
+ "epoch": 0.53,
274
+ "learning_rate": 0.00025775022143489815,
275
+ "loss": 0.9914,
276
+ "step": 420
277
+ },
278
+ {
279
+ "epoch": 0.55,
280
+ "learning_rate": 0.0002564216120460584,
281
+ "loss": 0.9924,
282
+ "step": 430
283
+ },
284
+ {
285
+ "epoch": 0.56,
286
+ "learning_rate": 0.00025509300265721876,
287
+ "loss": 0.9948,
288
+ "step": 440
289
+ },
290
+ {
291
+ "epoch": 0.57,
292
+ "learning_rate": 0.0002537643932683791,
293
+ "loss": 0.9879,
294
+ "step": 450
295
+ },
296
+ {
297
+ "epoch": 0.58,
298
+ "learning_rate": 0.0002524357838795394,
299
+ "loss": 0.988,
300
+ "step": 460
301
+ },
302
+ {
303
+ "epoch": 0.6,
304
+ "learning_rate": 0.0002511071744906997,
305
+ "loss": 0.9928,
306
+ "step": 470
307
+ },
308
+ {
309
+ "epoch": 0.61,
310
+ "learning_rate": 0.00024977856510186003,
311
+ "loss": 0.9916,
312
+ "step": 480
313
+ },
314
+ {
315
+ "epoch": 0.62,
316
+ "learning_rate": 0.00024844995571302037,
317
+ "loss": 0.994,
318
+ "step": 490
319
+ },
320
+ {
321
+ "epoch": 0.64,
322
+ "learning_rate": 0.0002471213463241807,
323
+ "loss": 0.996,
324
+ "step": 500
325
+ },
326
+ {
327
+ "epoch": 0.65,
328
+ "learning_rate": 0.000245792736935341,
329
+ "loss": 0.9783,
330
+ "step": 510
331
+ },
332
+ {
333
+ "epoch": 0.66,
334
+ "learning_rate": 0.0002444641275465013,
335
+ "loss": 0.9798,
336
+ "step": 520
337
+ },
338
+ {
339
+ "epoch": 0.67,
340
+ "learning_rate": 0.00024313551815766162,
341
+ "loss": 0.9791,
342
+ "step": 530
343
+ },
344
+ {
345
+ "epoch": 0.69,
346
+ "learning_rate": 0.00024180690876882195,
347
+ "loss": 0.9894,
348
+ "step": 540
349
+ },
350
+ {
351
+ "epoch": 0.7,
352
+ "learning_rate": 0.00024047829937998225,
353
+ "loss": 0.9904,
354
+ "step": 550
355
+ },
356
+ {
357
+ "epoch": 0.71,
358
+ "learning_rate": 0.00023914968999114256,
359
+ "loss": 0.9942,
360
+ "step": 560
361
+ },
362
+ {
363
+ "epoch": 0.72,
364
+ "learning_rate": 0.0002378210806023029,
365
+ "loss": 0.9897,
366
+ "step": 570
367
+ },
368
+ {
369
+ "epoch": 0.74,
370
+ "learning_rate": 0.0002364924712134632,
371
+ "loss": 0.9741,
372
+ "step": 580
373
+ },
374
+ {
375
+ "epoch": 0.75,
376
+ "learning_rate": 0.00023516386182462353,
377
+ "loss": 0.9631,
378
+ "step": 590
379
+ },
380
+ {
381
+ "epoch": 0.76,
382
+ "learning_rate": 0.00023383525243578384,
383
+ "loss": 0.9784,
384
+ "step": 600
385
+ },
386
+ {
387
+ "epoch": 0.76,
388
+ "eval_loss": 0.9769166111946106,
389
+ "eval_runtime": 78.8243,
390
+ "eval_samples_per_second": 25.373,
391
+ "eval_steps_per_second": 0.799,
392
+ "step": 600
393
+ },
394
+ {
395
+ "epoch": 0.78,
396
+ "learning_rate": 0.0002325066430469442,
397
+ "loss": 0.9703,
398
+ "step": 610
399
+ },
400
+ {
401
+ "epoch": 0.79,
402
+ "learning_rate": 0.00023117803365810448,
403
+ "loss": 0.9914,
404
+ "step": 620
405
+ },
406
+ {
407
+ "epoch": 0.8,
408
+ "learning_rate": 0.00022984942426926484,
409
+ "loss": 0.9522,
410
+ "step": 630
411
+ },
412
+ {
413
+ "epoch": 0.81,
414
+ "learning_rate": 0.00022852081488042514,
415
+ "loss": 0.9773,
416
+ "step": 640
417
+ },
418
+ {
419
+ "epoch": 0.83,
420
+ "learning_rate": 0.00022719220549158547,
421
+ "loss": 0.9954,
422
+ "step": 650
423
+ },
424
+ {
425
+ "epoch": 0.84,
426
+ "learning_rate": 0.00022586359610274578,
427
+ "loss": 0.9768,
428
+ "step": 660
429
+ },
430
+ {
431
+ "epoch": 0.85,
432
+ "learning_rate": 0.0002245349867139061,
433
+ "loss": 0.9908,
434
+ "step": 670
435
+ },
436
+ {
437
+ "epoch": 0.86,
438
+ "learning_rate": 0.00022320637732506642,
439
+ "loss": 0.9944,
440
+ "step": 680
441
+ },
442
+ {
443
+ "epoch": 0.88,
444
+ "learning_rate": 0.00022187776793622672,
445
+ "loss": 0.9772,
446
+ "step": 690
447
+ },
448
+ {
449
+ "epoch": 0.89,
450
+ "learning_rate": 0.00022054915854738706,
451
+ "loss": 0.9697,
452
+ "step": 700
453
+ },
454
+ {
455
+ "epoch": 0.9,
456
+ "learning_rate": 0.00021922054915854736,
457
+ "loss": 0.9652,
458
+ "step": 710
459
+ },
460
+ {
461
+ "epoch": 0.92,
462
+ "learning_rate": 0.0002178919397697077,
463
+ "loss": 0.9738,
464
+ "step": 720
465
+ },
466
+ {
467
+ "epoch": 0.93,
468
+ "learning_rate": 0.000216563330380868,
469
+ "loss": 0.9892,
470
+ "step": 730
471
+ },
472
+ {
473
+ "epoch": 0.94,
474
+ "learning_rate": 0.00021523472099202833,
475
+ "loss": 0.9731,
476
+ "step": 740
477
+ },
478
+ {
479
+ "epoch": 0.95,
480
+ "learning_rate": 0.00021390611160318864,
481
+ "loss": 0.985,
482
+ "step": 750
483
+ },
484
+ {
485
+ "epoch": 0.97,
486
+ "learning_rate": 0.00021257750221434897,
487
+ "loss": 0.9918,
488
+ "step": 760
489
+ },
490
+ {
491
+ "epoch": 0.98,
492
+ "learning_rate": 0.00021124889282550928,
493
+ "loss": 0.9603,
494
+ "step": 770
495
+ },
496
+ {
497
+ "epoch": 0.99,
498
+ "learning_rate": 0.0002099202834366696,
499
+ "loss": 0.9686,
500
+ "step": 780
501
+ },
502
+ {
503
+ "epoch": 1.0,
504
+ "learning_rate": 0.00020859167404782992,
505
+ "loss": 0.9761,
506
+ "step": 790
507
+ },
508
+ {
509
+ "epoch": 1.02,
510
+ "learning_rate": 0.00020726306465899025,
511
+ "loss": 0.9812,
512
+ "step": 800
513
+ },
514
+ {
515
+ "epoch": 1.02,
516
+ "eval_loss": 0.970382571220398,
517
+ "eval_runtime": 78.9078,
518
+ "eval_samples_per_second": 25.346,
519
+ "eval_steps_per_second": 0.798,
520
+ "step": 800
521
+ },
522
+ {
523
+ "epoch": 1.03,
524
+ "learning_rate": 0.00020593445527015056,
525
+ "loss": 0.95,
526
+ "step": 810
527
+ },
528
+ {
529
+ "epoch": 1.04,
530
+ "learning_rate": 0.0002046058458813109,
531
+ "loss": 0.9575,
532
+ "step": 820
533
+ },
534
+ {
535
+ "epoch": 1.06,
536
+ "learning_rate": 0.0002032772364924712,
537
+ "loss": 0.9826,
538
+ "step": 830
539
+ },
540
+ {
541
+ "epoch": 1.07,
542
+ "learning_rate": 0.0002019486271036315,
543
+ "loss": 0.973,
544
+ "step": 840
545
+ },
546
+ {
547
+ "epoch": 1.08,
548
+ "learning_rate": 0.00020062001771479183,
549
+ "loss": 0.9623,
550
+ "step": 850
551
+ },
552
+ {
553
+ "epoch": 1.09,
554
+ "learning_rate": 0.00019929140832595214,
555
+ "loss": 0.9618,
556
+ "step": 860
557
+ },
558
+ {
559
+ "epoch": 1.11,
560
+ "learning_rate": 0.00019796279893711247,
561
+ "loss": 0.9594,
562
+ "step": 870
563
+ },
564
+ {
565
+ "epoch": 1.12,
566
+ "learning_rate": 0.00019663418954827278,
567
+ "loss": 0.9827,
568
+ "step": 880
569
+ },
570
+ {
571
+ "epoch": 1.13,
572
+ "learning_rate": 0.0001953055801594331,
573
+ "loss": 0.9608,
574
+ "step": 890
575
+ },
576
+ {
577
+ "epoch": 1.14,
578
+ "learning_rate": 0.00019397697077059342,
579
+ "loss": 0.967,
580
+ "step": 900
581
+ },
582
+ {
583
+ "epoch": 1.16,
584
+ "learning_rate": 0.00019264836138175375,
585
+ "loss": 0.9498,
586
+ "step": 910
587
+ },
588
+ {
589
+ "epoch": 1.17,
590
+ "learning_rate": 0.00019131975199291405,
591
+ "loss": 0.972,
592
+ "step": 920
593
+ },
594
+ {
595
+ "epoch": 1.18,
596
+ "learning_rate": 0.0001899911426040744,
597
+ "loss": 0.9677,
598
+ "step": 930
599
+ },
600
+ {
601
+ "epoch": 1.2,
602
+ "learning_rate": 0.0001886625332152347,
603
+ "loss": 0.9715,
604
+ "step": 940
605
+ },
606
+ {
607
+ "epoch": 1.21,
608
+ "learning_rate": 0.00018733392382639503,
609
+ "loss": 0.9758,
610
+ "step": 950
611
+ },
612
+ {
613
+ "epoch": 1.22,
614
+ "learning_rate": 0.00018600531443755533,
615
+ "loss": 0.9667,
616
+ "step": 960
617
+ },
618
+ {
619
+ "epoch": 1.23,
620
+ "learning_rate": 0.00018467670504871564,
621
+ "loss": 0.9719,
622
+ "step": 970
623
+ },
624
+ {
625
+ "epoch": 1.25,
626
+ "learning_rate": 0.00018334809565987597,
627
+ "loss": 0.9584,
628
+ "step": 980
629
+ },
630
+ {
631
+ "epoch": 1.26,
632
+ "learning_rate": 0.00018201948627103628,
633
+ "loss": 0.9591,
634
+ "step": 990
635
+ },
636
+ {
637
+ "epoch": 1.27,
638
+ "learning_rate": 0.00018069087688219664,
639
+ "loss": 0.959,
640
+ "step": 1000
641
+ },
642
+ {
643
+ "epoch": 1.27,
644
+ "eval_loss": 0.9663413166999817,
645
+ "eval_runtime": 83.6093,
646
+ "eval_samples_per_second": 23.921,
647
+ "eval_steps_per_second": 0.754,
648
+ "step": 1000
649
+ },
650
+ {
651
+ "epoch": 1.28,
652
+ "learning_rate": 0.00017936226749335691,
653
+ "loss": 0.9545,
654
+ "step": 1010
655
+ },
656
+ {
657
+ "epoch": 1.3,
658
+ "learning_rate": 0.00017803365810451727,
659
+ "loss": 0.9675,
660
+ "step": 1020
661
+ },
662
+ {
663
+ "epoch": 1.31,
664
+ "learning_rate": 0.00017670504871567758,
665
+ "loss": 0.961,
666
+ "step": 1030
667
+ },
668
+ {
669
+ "epoch": 1.32,
670
+ "learning_rate": 0.0001753764393268379,
671
+ "loss": 0.9575,
672
+ "step": 1040
673
+ },
674
+ {
675
+ "epoch": 1.34,
676
+ "learning_rate": 0.00017404782993799822,
677
+ "loss": 0.9636,
678
+ "step": 1050
679
+ },
680
+ {
681
+ "epoch": 1.35,
682
+ "learning_rate": 0.00017271922054915855,
683
+ "loss": 0.9687,
684
+ "step": 1060
685
+ },
686
+ {
687
+ "epoch": 1.36,
688
+ "learning_rate": 0.00017139061116031886,
689
+ "loss": 0.9656,
690
+ "step": 1070
691
+ },
692
+ {
693
+ "epoch": 1.37,
694
+ "learning_rate": 0.0001700620017714792,
695
+ "loss": 0.9493,
696
+ "step": 1080
697
+ },
698
+ {
699
+ "epoch": 1.39,
700
+ "learning_rate": 0.0001687333923826395,
701
+ "loss": 0.9681,
702
+ "step": 1090
703
+ },
704
+ {
705
+ "epoch": 1.4,
706
+ "learning_rate": 0.00016740478299379983,
707
+ "loss": 0.9537,
708
+ "step": 1100
709
+ },
710
+ {
711
+ "epoch": 1.41,
712
+ "learning_rate": 0.00016607617360496013,
713
+ "loss": 0.9468,
714
+ "step": 1110
715
+ },
716
+ {
717
+ "epoch": 1.42,
718
+ "learning_rate": 0.00016474756421612044,
719
+ "loss": 0.957,
720
+ "step": 1120
721
+ },
722
+ {
723
+ "epoch": 1.44,
724
+ "learning_rate": 0.00016341895482728077,
725
+ "loss": 0.9697,
726
+ "step": 1130
727
+ },
728
+ {
729
+ "epoch": 1.45,
730
+ "learning_rate": 0.00016209034543844108,
731
+ "loss": 0.9689,
732
+ "step": 1140
733
+ },
734
+ {
735
+ "epoch": 1.46,
736
+ "learning_rate": 0.0001607617360496014,
737
+ "loss": 0.9691,
738
+ "step": 1150
739
+ },
740
+ {
741
+ "epoch": 1.47,
742
+ "learning_rate": 0.00015943312666076172,
743
+ "loss": 0.9678,
744
+ "step": 1160
745
+ },
746
+ {
747
+ "epoch": 1.49,
748
+ "learning_rate": 0.00015810451727192205,
749
+ "loss": 0.967,
750
+ "step": 1170
751
+ },
752
+ {
753
+ "epoch": 1.5,
754
+ "learning_rate": 0.00015677590788308235,
755
+ "loss": 0.9742,
756
+ "step": 1180
757
+ },
758
+ {
759
+ "epoch": 1.51,
760
+ "learning_rate": 0.0001554472984942427,
761
+ "loss": 0.9438,
762
+ "step": 1190
763
+ },
764
+ {
765
+ "epoch": 1.53,
766
+ "learning_rate": 0.000154118689105403,
767
+ "loss": 0.9688,
768
+ "step": 1200
769
+ },
770
+ {
771
+ "epoch": 1.53,
772
+ "eval_loss": 0.9622219204902649,
773
+ "eval_runtime": 78.9396,
774
+ "eval_samples_per_second": 25.336,
775
+ "eval_steps_per_second": 0.798,
776
+ "step": 1200
777
+ },
778
+ {
779
+ "epoch": 1.54,
780
+ "learning_rate": 0.00015279007971656333,
781
+ "loss": 0.9655,
782
+ "step": 1210
783
+ },
784
+ {
785
+ "epoch": 1.55,
786
+ "learning_rate": 0.00015146147032772363,
787
+ "loss": 0.957,
788
+ "step": 1220
789
+ },
790
+ {
791
+ "epoch": 1.56,
792
+ "learning_rate": 0.00015013286093888396,
793
+ "loss": 0.9667,
794
+ "step": 1230
795
+ },
796
+ {
797
+ "epoch": 1.58,
798
+ "learning_rate": 0.00014880425155004427,
799
+ "loss": 0.9554,
800
+ "step": 1240
801
+ },
802
+ {
803
+ "epoch": 1.59,
804
+ "learning_rate": 0.0001474756421612046,
805
+ "loss": 0.9691,
806
+ "step": 1250
807
+ },
808
+ {
809
+ "epoch": 1.6,
810
+ "learning_rate": 0.0001461470327723649,
811
+ "loss": 0.9587,
812
+ "step": 1260
813
+ },
814
+ {
815
+ "epoch": 1.61,
816
+ "learning_rate": 0.00014481842338352524,
817
+ "loss": 0.9461,
818
+ "step": 1270
819
+ },
820
+ {
821
+ "epoch": 1.63,
822
+ "learning_rate": 0.00014348981399468555,
823
+ "loss": 0.9607,
824
+ "step": 1280
825
+ },
826
+ {
827
+ "epoch": 1.64,
828
+ "learning_rate": 0.00014216120460584588,
829
+ "loss": 0.9528,
830
+ "step": 1290
831
+ },
832
+ {
833
+ "epoch": 1.65,
834
+ "learning_rate": 0.00014083259521700619,
835
+ "loss": 0.9469,
836
+ "step": 1300
837
+ },
838
+ {
839
+ "epoch": 1.67,
840
+ "learning_rate": 0.00013950398582816652,
841
+ "loss": 0.9415,
842
+ "step": 1310
843
+ },
844
+ {
845
+ "epoch": 1.68,
846
+ "learning_rate": 0.00013817537643932682,
847
+ "loss": 0.9628,
848
+ "step": 1320
849
+ },
850
+ {
851
+ "epoch": 1.69,
852
+ "learning_rate": 0.00013684676705048713,
853
+ "loss": 0.9562,
854
+ "step": 1330
855
+ },
856
+ {
857
+ "epoch": 1.7,
858
+ "learning_rate": 0.00013551815766164746,
859
+ "loss": 0.9579,
860
+ "step": 1340
861
+ },
862
+ {
863
+ "epoch": 1.72,
864
+ "learning_rate": 0.00013418954827280777,
865
+ "loss": 0.9578,
866
+ "step": 1350
867
+ },
868
+ {
869
+ "epoch": 1.73,
870
+ "learning_rate": 0.0001328609388839681,
871
+ "loss": 0.9508,
872
+ "step": 1360
873
+ },
874
+ {
875
+ "epoch": 1.74,
876
+ "learning_rate": 0.0001315323294951284,
877
+ "loss": 0.9596,
878
+ "step": 1370
879
+ },
880
+ {
881
+ "epoch": 1.75,
882
+ "learning_rate": 0.00013020372010628874,
883
+ "loss": 0.9446,
884
+ "step": 1380
885
+ },
886
+ {
887
+ "epoch": 1.77,
888
+ "learning_rate": 0.00012887511071744907,
889
+ "loss": 0.9418,
890
+ "step": 1390
891
+ },
892
+ {
893
+ "epoch": 1.78,
894
+ "learning_rate": 0.00012754650132860938,
895
+ "loss": 0.9447,
896
+ "step": 1400
897
+ },
898
+ {
899
+ "epoch": 1.78,
900
+ "eval_loss": 0.9596577286720276,
901
+ "eval_runtime": 78.8874,
902
+ "eval_samples_per_second": 25.353,
903
+ "eval_steps_per_second": 0.799,
904
+ "step": 1400
905
+ },
906
+ {
907
+ "epoch": 1.79,
908
+ "learning_rate": 0.0001262178919397697,
909
+ "loss": 0.9597,
910
+ "step": 1410
911
+ },
912
+ {
913
+ "epoch": 1.81,
914
+ "learning_rate": 0.00012488928255093002,
915
+ "loss": 0.9601,
916
+ "step": 1420
917
+ },
918
+ {
919
+ "epoch": 1.82,
920
+ "learning_rate": 0.00012356067316209035,
921
+ "loss": 0.9794,
922
+ "step": 1430
923
+ },
924
+ {
925
+ "epoch": 1.83,
926
+ "learning_rate": 0.00012223206377325066,
927
+ "loss": 0.9295,
928
+ "step": 1440
929
+ },
930
+ {
931
+ "epoch": 1.84,
932
+ "learning_rate": 0.00012090345438441097,
933
+ "loss": 0.9418,
934
+ "step": 1450
935
+ },
936
+ {
937
+ "epoch": 1.86,
938
+ "learning_rate": 0.00011957484499557128,
939
+ "loss": 0.941,
940
+ "step": 1460
941
+ },
942
+ {
943
+ "epoch": 1.87,
944
+ "learning_rate": 0.0001182462356067316,
945
+ "loss": 0.9504,
946
+ "step": 1470
947
+ },
948
+ {
949
+ "epoch": 1.88,
950
+ "learning_rate": 0.00011691762621789192,
951
+ "loss": 0.9489,
952
+ "step": 1480
953
+ },
954
+ {
955
+ "epoch": 1.89,
956
+ "learning_rate": 0.00011558901682905224,
957
+ "loss": 0.9482,
958
+ "step": 1490
959
+ },
960
+ {
961
+ "epoch": 1.91,
962
+ "learning_rate": 0.00011426040744021257,
963
+ "loss": 0.9435,
964
+ "step": 1500
965
+ },
966
+ {
967
+ "epoch": 1.92,
968
+ "learning_rate": 0.00011293179805137289,
969
+ "loss": 0.953,
970
+ "step": 1510
971
+ },
972
+ {
973
+ "epoch": 1.93,
974
+ "learning_rate": 0.00011160318866253321,
975
+ "loss": 0.9468,
976
+ "step": 1520
977
+ },
978
+ {
979
+ "epoch": 1.95,
980
+ "learning_rate": 0.00011027457927369353,
981
+ "loss": 0.9289,
982
+ "step": 1530
983
+ },
984
+ {
985
+ "epoch": 1.96,
986
+ "learning_rate": 0.00010894596988485385,
987
+ "loss": 0.9533,
988
+ "step": 1540
989
+ },
990
+ {
991
+ "epoch": 1.97,
992
+ "learning_rate": 0.00010761736049601417,
993
+ "loss": 0.9591,
994
+ "step": 1550
995
+ },
996
+ {
997
+ "epoch": 1.98,
998
+ "learning_rate": 0.00010628875110717449,
999
+ "loss": 0.951,
1000
+ "step": 1560
1001
+ },
1002
+ {
1003
+ "epoch": 2.0,
1004
+ "learning_rate": 0.0001049601417183348,
1005
+ "loss": 0.9171,
1006
+ "step": 1570
1007
+ },
1008
+ {
1009
+ "epoch": 2.01,
1010
+ "learning_rate": 0.00010363153232949512,
1011
+ "loss": 0.9375,
1012
+ "step": 1580
1013
+ },
1014
+ {
1015
+ "epoch": 2.02,
1016
+ "learning_rate": 0.00010230292294065544,
1017
+ "loss": 0.9568,
1018
+ "step": 1590
1019
+ },
1020
+ {
1021
+ "epoch": 2.03,
1022
+ "learning_rate": 0.00010097431355181575,
1023
+ "loss": 0.9633,
1024
+ "step": 1600
1025
+ },
1026
+ {
1027
+ "epoch": 2.03,
1028
+ "eval_loss": 0.9573116302490234,
1029
+ "eval_runtime": 79.0309,
1030
+ "eval_samples_per_second": 25.307,
1031
+ "eval_steps_per_second": 0.797,
1032
+ "step": 1600
1033
+ },
1034
+ {
1035
+ "epoch": 2.05,
1036
+ "learning_rate": 9.964570416297607e-05,
1037
+ "loss": 0.9424,
1038
+ "step": 1610
1039
+ },
1040
+ {
1041
+ "epoch": 2.06,
1042
+ "learning_rate": 9.831709477413639e-05,
1043
+ "loss": 0.9243,
1044
+ "step": 1620
1045
+ },
1046
+ {
1047
+ "epoch": 2.07,
1048
+ "learning_rate": 9.698848538529671e-05,
1049
+ "loss": 0.9415,
1050
+ "step": 1630
1051
+ },
1052
+ {
1053
+ "epoch": 2.09,
1054
+ "learning_rate": 9.565987599645703e-05,
1055
+ "loss": 0.9434,
1056
+ "step": 1640
1057
+ },
1058
+ {
1059
+ "epoch": 2.1,
1060
+ "learning_rate": 9.433126660761735e-05,
1061
+ "loss": 0.9467,
1062
+ "step": 1650
1063
+ },
1064
+ {
1065
+ "epoch": 2.11,
1066
+ "learning_rate": 9.300265721877767e-05,
1067
+ "loss": 0.9516,
1068
+ "step": 1660
1069
+ },
1070
+ {
1071
+ "epoch": 2.12,
1072
+ "learning_rate": 9.167404782993798e-05,
1073
+ "loss": 0.9617,
1074
+ "step": 1670
1075
+ },
1076
+ {
1077
+ "epoch": 2.14,
1078
+ "learning_rate": 9.034543844109832e-05,
1079
+ "loss": 0.9379,
1080
+ "step": 1680
1081
+ },
1082
+ {
1083
+ "epoch": 2.15,
1084
+ "learning_rate": 8.901682905225864e-05,
1085
+ "loss": 0.9455,
1086
+ "step": 1690
1087
+ },
1088
+ {
1089
+ "epoch": 2.16,
1090
+ "learning_rate": 8.768821966341896e-05,
1091
+ "loss": 0.9477,
1092
+ "step": 1700
1093
+ },
1094
+ {
1095
+ "epoch": 2.17,
1096
+ "learning_rate": 8.635961027457928e-05,
1097
+ "loss": 0.9448,
1098
+ "step": 1710
1099
+ },
1100
+ {
1101
+ "epoch": 2.19,
1102
+ "learning_rate": 8.50310008857396e-05,
1103
+ "loss": 0.9361,
1104
+ "step": 1720
1105
+ },
1106
+ {
1107
+ "epoch": 2.2,
1108
+ "learning_rate": 8.370239149689991e-05,
1109
+ "loss": 0.922,
1110
+ "step": 1730
1111
+ },
1112
+ {
1113
+ "epoch": 2.21,
1114
+ "learning_rate": 8.237378210806022e-05,
1115
+ "loss": 0.9442,
1116
+ "step": 1740
1117
+ },
1118
+ {
1119
+ "epoch": 2.23,
1120
+ "learning_rate": 8.104517271922054e-05,
1121
+ "loss": 0.9575,
1122
+ "step": 1750
1123
+ },
1124
+ {
1125
+ "epoch": 2.24,
1126
+ "learning_rate": 7.971656333038086e-05,
1127
+ "loss": 0.929,
1128
+ "step": 1760
1129
+ },
1130
+ {
1131
+ "epoch": 2.25,
1132
+ "learning_rate": 7.838795394154118e-05,
1133
+ "loss": 0.9312,
1134
+ "step": 1770
1135
+ },
1136
+ {
1137
+ "epoch": 2.26,
1138
+ "learning_rate": 7.70593445527015e-05,
1139
+ "loss": 0.943,
1140
+ "step": 1780
1141
+ },
1142
+ {
1143
+ "epoch": 2.28,
1144
+ "learning_rate": 7.573073516386182e-05,
1145
+ "loss": 0.9416,
1146
+ "step": 1790
1147
+ },
1148
+ {
1149
+ "epoch": 2.29,
1150
+ "learning_rate": 7.440212577502214e-05,
1151
+ "loss": 0.9455,
1152
+ "step": 1800
1153
+ },
1154
+ {
1155
+ "epoch": 2.29,
1156
+ "eval_loss": 0.9558648467063904,
1157
+ "eval_runtime": 79.0765,
1158
+ "eval_samples_per_second": 25.292,
1159
+ "eval_steps_per_second": 0.797,
1160
+ "step": 1800
1161
+ },
1162
+ {
1163
+ "epoch": 2.3,
1164
+ "learning_rate": 7.307351638618245e-05,
1165
+ "loss": 0.9307,
1166
+ "step": 1810
1167
+ },
1168
+ {
1169
+ "epoch": 2.31,
1170
+ "learning_rate": 7.174490699734277e-05,
1171
+ "loss": 0.9517,
1172
+ "step": 1820
1173
+ },
1174
+ {
1175
+ "epoch": 2.33,
1176
+ "learning_rate": 7.041629760850309e-05,
1177
+ "loss": 0.9298,
1178
+ "step": 1830
1179
+ },
1180
+ {
1181
+ "epoch": 2.34,
1182
+ "learning_rate": 6.908768821966341e-05,
1183
+ "loss": 0.936,
1184
+ "step": 1840
1185
+ },
1186
+ {
1187
+ "epoch": 2.35,
1188
+ "learning_rate": 6.775907883082373e-05,
1189
+ "loss": 0.944,
1190
+ "step": 1850
1191
+ },
1192
+ {
1193
+ "epoch": 2.36,
1194
+ "learning_rate": 6.643046944198405e-05,
1195
+ "loss": 0.9521,
1196
+ "step": 1860
1197
+ },
1198
+ {
1199
+ "epoch": 2.38,
1200
+ "learning_rate": 6.510186005314437e-05,
1201
+ "loss": 0.9579,
1202
+ "step": 1870
1203
+ },
1204
+ {
1205
+ "epoch": 2.39,
1206
+ "learning_rate": 6.377325066430469e-05,
1207
+ "loss": 0.9412,
1208
+ "step": 1880
1209
+ },
1210
+ {
1211
+ "epoch": 2.4,
1212
+ "learning_rate": 6.244464127546501e-05,
1213
+ "loss": 0.9268,
1214
+ "step": 1890
1215
+ },
1216
+ {
1217
+ "epoch": 2.42,
1218
+ "learning_rate": 6.111603188662533e-05,
1219
+ "loss": 0.9343,
1220
+ "step": 1900
1221
+ },
1222
+ {
1223
+ "epoch": 2.43,
1224
+ "learning_rate": 5.978742249778564e-05,
1225
+ "loss": 0.9517,
1226
+ "step": 1910
1227
+ },
1228
+ {
1229
+ "epoch": 2.44,
1230
+ "learning_rate": 5.845881310894596e-05,
1231
+ "loss": 0.944,
1232
+ "step": 1920
1233
+ },
1234
+ {
1235
+ "epoch": 2.45,
1236
+ "learning_rate": 5.7130203720106285e-05,
1237
+ "loss": 0.9384,
1238
+ "step": 1930
1239
+ },
1240
+ {
1241
+ "epoch": 2.47,
1242
+ "learning_rate": 5.5801594331266605e-05,
1243
+ "loss": 0.9462,
1244
+ "step": 1940
1245
+ },
1246
+ {
1247
+ "epoch": 2.48,
1248
+ "learning_rate": 5.4472984942426924e-05,
1249
+ "loss": 0.9317,
1250
+ "step": 1950
1251
+ },
1252
+ {
1253
+ "epoch": 2.49,
1254
+ "learning_rate": 5.314437555358724e-05,
1255
+ "loss": 0.9354,
1256
+ "step": 1960
1257
+ },
1258
+ {
1259
+ "epoch": 2.5,
1260
+ "learning_rate": 5.181576616474756e-05,
1261
+ "loss": 0.9261,
1262
+ "step": 1970
1263
+ },
1264
+ {
1265
+ "epoch": 2.52,
1266
+ "learning_rate": 5.0487156775907875e-05,
1267
+ "loss": 0.9317,
1268
+ "step": 1980
1269
+ },
1270
+ {
1271
+ "epoch": 2.53,
1272
+ "learning_rate": 4.9158547387068194e-05,
1273
+ "loss": 0.932,
1274
+ "step": 1990
1275
+ },
1276
+ {
1277
+ "epoch": 2.54,
1278
+ "learning_rate": 4.7829937998228513e-05,
1279
+ "loss": 0.9323,
1280
+ "step": 2000
1281
+ },
1282
+ {
1283
+ "epoch": 2.54,
1284
+ "eval_loss": 0.9545624852180481,
1285
+ "eval_runtime": 78.9727,
1286
+ "eval_samples_per_second": 25.325,
1287
+ "eval_steps_per_second": 0.798,
1288
+ "step": 2000
1289
+ },
1290
+ {
1291
+ "epoch": 2.56,
1292
+ "learning_rate": 4.650132860938883e-05,
1293
+ "loss": 0.9337,
1294
+ "step": 2010
1295
+ },
1296
+ {
1297
+ "epoch": 2.57,
1298
+ "learning_rate": 4.517271922054916e-05,
1299
+ "loss": 0.9304,
1300
+ "step": 2020
1301
+ },
1302
+ {
1303
+ "epoch": 2.58,
1304
+ "learning_rate": 4.384410983170948e-05,
1305
+ "loss": 0.9119,
1306
+ "step": 2030
1307
+ },
1308
+ {
1309
+ "epoch": 2.59,
1310
+ "learning_rate": 4.25155004428698e-05,
1311
+ "loss": 0.9413,
1312
+ "step": 2040
1313
+ },
1314
+ {
1315
+ "epoch": 2.61,
1316
+ "learning_rate": 4.118689105403011e-05,
1317
+ "loss": 0.9462,
1318
+ "step": 2050
1319
+ },
1320
+ {
1321
+ "epoch": 2.62,
1322
+ "learning_rate": 3.985828166519043e-05,
1323
+ "loss": 0.9459,
1324
+ "step": 2060
1325
+ },
1326
+ {
1327
+ "epoch": 2.63,
1328
+ "learning_rate": 3.852967227635075e-05,
1329
+ "loss": 0.944,
1330
+ "step": 2070
1331
+ },
1332
+ {
1333
+ "epoch": 2.64,
1334
+ "learning_rate": 3.720106288751107e-05,
1335
+ "loss": 0.9436,
1336
+ "step": 2080
1337
+ },
1338
+ {
1339
+ "epoch": 2.66,
1340
+ "learning_rate": 3.587245349867139e-05,
1341
+ "loss": 0.9286,
1342
+ "step": 2090
1343
+ },
1344
+ {
1345
+ "epoch": 2.67,
1346
+ "learning_rate": 3.4543844109831706e-05,
1347
+ "loss": 0.9352,
1348
+ "step": 2100
1349
+ },
1350
+ {
1351
+ "epoch": 2.68,
1352
+ "learning_rate": 3.3215234720992025e-05,
1353
+ "loss": 0.9363,
1354
+ "step": 2110
1355
+ },
1356
+ {
1357
+ "epoch": 2.7,
1358
+ "learning_rate": 3.1886625332152345e-05,
1359
+ "loss": 0.9498,
1360
+ "step": 2120
1361
+ },
1362
+ {
1363
+ "epoch": 2.71,
1364
+ "learning_rate": 3.0558015943312664e-05,
1365
+ "loss": 0.9392,
1366
+ "step": 2130
1367
+ },
1368
+ {
1369
+ "epoch": 2.72,
1370
+ "learning_rate": 2.922940655447298e-05,
1371
+ "loss": 0.9429,
1372
+ "step": 2140
1373
+ },
1374
+ {
1375
+ "epoch": 2.73,
1376
+ "learning_rate": 2.7900797165633302e-05,
1377
+ "loss": 0.9469,
1378
+ "step": 2150
1379
+ },
1380
+ {
1381
+ "epoch": 2.75,
1382
+ "learning_rate": 2.657218777679362e-05,
1383
+ "loss": 0.9452,
1384
+ "step": 2160
1385
+ },
1386
+ {
1387
+ "epoch": 2.76,
1388
+ "learning_rate": 2.5243578387953937e-05,
1389
+ "loss": 0.9215,
1390
+ "step": 2170
1391
+ },
1392
+ {
1393
+ "epoch": 2.77,
1394
+ "learning_rate": 2.3914968999114257e-05,
1395
+ "loss": 0.932,
1396
+ "step": 2180
1397
+ },
1398
+ {
1399
+ "epoch": 2.78,
1400
+ "learning_rate": 2.258635961027458e-05,
1401
+ "loss": 0.931,
1402
+ "step": 2190
1403
+ },
1404
+ {
1405
+ "epoch": 2.8,
1406
+ "learning_rate": 2.12577502214349e-05,
1407
+ "loss": 0.9507,
1408
+ "step": 2200
1409
+ },
1410
+ {
1411
+ "epoch": 2.8,
1412
+ "eval_loss": 0.953123152256012,
1413
+ "eval_runtime": 78.9977,
1414
+ "eval_samples_per_second": 25.317,
1415
+ "eval_steps_per_second": 0.797,
1416
+ "step": 2200
1417
+ }
1418
+ ],
1419
+ "max_steps": 2358,
1420
+ "num_train_epochs": 3,
1421
+ "total_flos": 7.228289393478337e+18,
1422
+ "trial_name": null,
1423
+ "trial_params": null
1424
+ }
checkpoint-2200/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d466c45720cc9bfdf9011921a4cde00ee73d739fae921d7ae51485f2cfe2556d
3
+ size 3579