File size: 9,385 Bytes
daeda6a
a784dc8
 
 
daeda6a
a784dc8
daeda6a
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
0d00a7b
 
 
 
 
 
 
 
 
 
 
 
 
 
91a648a
 
 
 
 
 
 
 
 
 
 
 
 
 
269e4c8
 
 
 
 
 
 
 
 
 
 
 
 
 
1f67f12
 
 
 
 
 
 
 
 
 
 
 
 
 
c55c7e2
 
 
 
 
 
 
 
 
 
 
 
 
 
4ed2b64
 
 
 
 
 
 
 
 
 
 
 
 
 
399fcbd
 
 
 
 
 
 
 
 
 
 
 
 
 
d046ad8
 
 
 
 
 
 
 
 
 
 
 
 
 
68249f1
 
 
 
 
 
 
 
 
 
 
 
 
 
ba44396
 
 
 
a784dc8
ba44396
 
 
 
a784dc8
 
 
 
ba44396
d5c6da3
 
 
 
a784dc8
d5c6da3
 
 
 
a784dc8
 
 
 
d5c6da3
8edc9b8
 
 
 
a784dc8
8edc9b8
 
 
 
a784dc8
 
 
 
8edc9b8
52341af
 
 
 
a784dc8
52341af
 
 
 
a784dc8
 
 
 
52341af
edb64aa
 
 
 
a784dc8
edb64aa
 
 
 
a784dc8
 
 
 
edb64aa
9bf2345
 
 
 
a784dc8
9bf2345
 
 
 
a784dc8
 
 
 
9bf2345
5bddf13
 
 
 
a784dc8
5bddf13
 
 
 
a784dc8
 
 
 
5bddf13
c982031
 
 
 
a784dc8
c982031
 
 
 
a784dc8
 
 
 
c982031
2c460a0
 
 
 
a784dc8
2c460a0
 
 
 
a784dc8
 
 
 
2c460a0
f052bbf
 
 
 
a784dc8
f052bbf
 
 
 
a784dc8
 
 
 
f052bbf
0be86e2
 
 
 
a784dc8
0be86e2
 
 
 
a784dc8
 
 
 
0be86e2
b9c05f4
 
 
 
a784dc8
b9c05f4
 
 
 
a784dc8
 
 
 
b9c05f4
7222fbd
 
 
 
a784dc8
7222fbd
 
 
 
a784dc8
 
 
 
7222fbd
517418a
 
 
 
a784dc8
517418a
 
 
 
a784dc8
 
 
 
517418a
741aef9
 
 
 
a784dc8
741aef9
 
 
 
a784dc8
 
 
 
741aef9
c8b3ff0
 
 
 
a784dc8
c8b3ff0
 
 
 
a784dc8
 
 
 
c8b3ff0
3978269
 
 
 
a784dc8
3978269
 
 
 
a784dc8
 
 
 
3978269
99c32fe
 
 
 
a784dc8
99c32fe
 
 
 
a784dc8
 
 
 
99c32fe
28764c0
 
 
 
a784dc8
28764c0
 
 
 
a784dc8
 
 
 
28764c0
daeda6a
 
 
 
 
 
a784dc8
daeda6a
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
{
  "best_metric": 1.333436369895935,
  "best_model_checkpoint": "./outputs/checkpoint-2900",
  "epoch": 2.1129326047358834,
  "eval_steps": 100,
  "global_step": 2900,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 2.182,
      "step": 100
    },
    {
      "epoch": 0.07,
      "eval_loss": 2.0125839710235596,
      "eval_runtime": 144.1418,
      "eval_samples_per_second": 43.527,
      "eval_steps_per_second": 5.446,
      "step": 100
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 1.9626,
      "step": 200
    },
    {
      "epoch": 0.15,
      "eval_loss": 1.9195271730422974,
      "eval_runtime": 144.2253,
      "eval_samples_per_second": 43.501,
      "eval_steps_per_second": 5.443,
      "step": 200
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 1.8881,
      "step": 300
    },
    {
      "epoch": 0.22,
      "eval_loss": 1.858821988105774,
      "eval_runtime": 144.0313,
      "eval_samples_per_second": 43.56,
      "eval_steps_per_second": 5.45,
      "step": 300
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 1.8367,
      "step": 400
    },
    {
      "epoch": 0.29,
      "eval_loss": 1.8125503063201904,
      "eval_runtime": 143.9861,
      "eval_samples_per_second": 43.574,
      "eval_steps_per_second": 5.452,
      "step": 400
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 1.785,
      "step": 500
    },
    {
      "epoch": 0.36,
      "eval_loss": 1.7720445394515991,
      "eval_runtime": 144.0061,
      "eval_samples_per_second": 43.568,
      "eval_steps_per_second": 5.451,
      "step": 500
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 1.7562,
      "step": 600
    },
    {
      "epoch": 0.44,
      "eval_loss": 1.7424511909484863,
      "eval_runtime": 144.0245,
      "eval_samples_per_second": 43.562,
      "eval_steps_per_second": 5.45,
      "step": 600
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 1.7176,
      "step": 700
    },
    {
      "epoch": 0.51,
      "eval_loss": 1.7112771272659302,
      "eval_runtime": 144.2204,
      "eval_samples_per_second": 43.503,
      "eval_steps_per_second": 5.443,
      "step": 700
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002,
      "loss": 1.6969,
      "step": 800
    },
    {
      "epoch": 0.58,
      "eval_loss": 1.6823335886001587,
      "eval_runtime": 144.0134,
      "eval_samples_per_second": 43.565,
      "eval_steps_per_second": 5.451,
      "step": 800
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002,
      "loss": 1.6619,
      "step": 900
    },
    {
      "epoch": 0.66,
      "eval_loss": 1.6567240953445435,
      "eval_runtime": 144.0655,
      "eval_samples_per_second": 43.55,
      "eval_steps_per_second": 5.449,
      "step": 900
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 1.6469,
      "step": 1000
    },
    {
      "epoch": 0.73,
      "eval_loss": 1.6352009773254395,
      "eval_runtime": 144.0664,
      "eval_samples_per_second": 43.549,
      "eval_steps_per_second": 5.449,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002,
      "loss": 1.6461,
      "step": 1100
    },
    {
      "epoch": 0.8,
      "eval_loss": 1.6240431070327759,
      "eval_runtime": 293.5989,
      "eval_samples_per_second": 21.369,
      "eval_steps_per_second": 2.674,
      "step": 1100
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002,
      "loss": 1.5992,
      "step": 1200
    },
    {
      "epoch": 0.87,
      "eval_loss": 1.5974311828613281,
      "eval_runtime": 291.7,
      "eval_samples_per_second": 21.508,
      "eval_steps_per_second": 2.691,
      "step": 1200
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002,
      "loss": 1.6021,
      "step": 1300
    },
    {
      "epoch": 0.95,
      "eval_loss": 1.5751127004623413,
      "eval_runtime": 289.9524,
      "eval_samples_per_second": 21.638,
      "eval_steps_per_second": 2.707,
      "step": 1300
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0002,
      "loss": 1.5538,
      "step": 1400
    },
    {
      "epoch": 1.02,
      "eval_loss": 1.5539450645446777,
      "eval_runtime": 287.8748,
      "eval_samples_per_second": 21.794,
      "eval_steps_per_second": 2.727,
      "step": 1400
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002,
      "loss": 1.5249,
      "step": 1500
    },
    {
      "epoch": 1.09,
      "eval_loss": 1.5348094701766968,
      "eval_runtime": 287.891,
      "eval_samples_per_second": 21.793,
      "eval_steps_per_second": 2.727,
      "step": 1500
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002,
      "loss": 1.506,
      "step": 1600
    },
    {
      "epoch": 1.17,
      "eval_loss": 1.515953540802002,
      "eval_runtime": 289.836,
      "eval_samples_per_second": 21.647,
      "eval_steps_per_second": 2.708,
      "step": 1600
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002,
      "loss": 1.5042,
      "step": 1700
    },
    {
      "epoch": 1.24,
      "eval_loss": 1.4988901615142822,
      "eval_runtime": 291.5471,
      "eval_samples_per_second": 21.52,
      "eval_steps_per_second": 2.693,
      "step": 1700
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002,
      "loss": 1.4762,
      "step": 1800
    },
    {
      "epoch": 1.31,
      "eval_loss": 1.4844294786453247,
      "eval_runtime": 293.6668,
      "eval_samples_per_second": 21.364,
      "eval_steps_per_second": 2.673,
      "step": 1800
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002,
      "loss": 1.4652,
      "step": 1900
    },
    {
      "epoch": 1.38,
      "eval_loss": 1.4694663286209106,
      "eval_runtime": 295.5867,
      "eval_samples_per_second": 21.226,
      "eval_steps_per_second": 2.656,
      "step": 1900
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0002,
      "loss": 1.4481,
      "step": 2000
    },
    {
      "epoch": 1.46,
      "eval_loss": 1.4534634351730347,
      "eval_runtime": 296.7451,
      "eval_samples_per_second": 21.143,
      "eval_steps_per_second": 2.645,
      "step": 2000
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002,
      "loss": 1.4335,
      "step": 2100
    },
    {
      "epoch": 1.53,
      "eval_loss": 1.4383305311203003,
      "eval_runtime": 294.659,
      "eval_samples_per_second": 21.292,
      "eval_steps_per_second": 2.664,
      "step": 2100
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002,
      "loss": 1.4075,
      "step": 2200
    },
    {
      "epoch": 1.6,
      "eval_loss": 1.4232139587402344,
      "eval_runtime": 292.737,
      "eval_samples_per_second": 21.432,
      "eval_steps_per_second": 2.682,
      "step": 2200
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002,
      "loss": 1.4059,
      "step": 2300
    },
    {
      "epoch": 1.68,
      "eval_loss": 1.411597490310669,
      "eval_runtime": 290.6759,
      "eval_samples_per_second": 21.584,
      "eval_steps_per_second": 2.701,
      "step": 2300
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0002,
      "loss": 1.4055,
      "step": 2400
    },
    {
      "epoch": 1.75,
      "eval_loss": 1.3974188566207886,
      "eval_runtime": 288.6807,
      "eval_samples_per_second": 21.733,
      "eval_steps_per_second": 2.719,
      "step": 2400
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0002,
      "loss": 1.3772,
      "step": 2500
    },
    {
      "epoch": 1.82,
      "eval_loss": 1.384261965751648,
      "eval_runtime": 287.1594,
      "eval_samples_per_second": 21.848,
      "eval_steps_per_second": 2.734,
      "step": 2500
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0002,
      "loss": 1.3669,
      "step": 2600
    },
    {
      "epoch": 1.89,
      "eval_loss": 1.3715009689331055,
      "eval_runtime": 288.6305,
      "eval_samples_per_second": 21.737,
      "eval_steps_per_second": 2.72,
      "step": 2600
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0002,
      "loss": 1.3548,
      "step": 2700
    },
    {
      "epoch": 1.97,
      "eval_loss": 1.3575737476348877,
      "eval_runtime": 290.5389,
      "eval_samples_per_second": 21.594,
      "eval_steps_per_second": 2.702,
      "step": 2700
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0002,
      "loss": 1.3266,
      "step": 2800
    },
    {
      "epoch": 2.04,
      "eval_loss": 1.3451271057128906,
      "eval_runtime": 292.4987,
      "eval_samples_per_second": 21.45,
      "eval_steps_per_second": 2.684,
      "step": 2800
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0002,
      "loss": 1.3004,
      "step": 2900
    },
    {
      "epoch": 2.11,
      "eval_loss": 1.333436369895935,
      "eval_runtime": 294.3881,
      "eval_samples_per_second": 21.312,
      "eval_steps_per_second": 2.667,
      "step": 2900
    }
  ],
  "logging_steps": 100,
  "max_steps": 4116,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 1.72867124009558e+17,
  "trial_name": null,
  "trial_params": null
}