File size: 9,014 Bytes
de2836b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.4032768978700165,
  "eval_steps": 500,
  "global_step": 11000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.10922992900054615,
      "grad_norm": 0.9561710357666016,
      "learning_rate": 1.4564520827264783e-05,
      "loss": 0.7634,
      "step": 500
    },
    {
      "epoch": 0.10922992900054615,
      "eval_loss": 0.3866455852985382,
      "eval_runtime": 53.8542,
      "eval_samples_per_second": 1.727,
      "eval_steps_per_second": 1.727,
      "step": 500
    },
    {
      "epoch": 0.2184598580010923,
      "grad_norm": 0.6381384134292603,
      "learning_rate": 2.9129041654529566e-05,
      "loss": 0.2265,
      "step": 1000
    },
    {
      "epoch": 0.2184598580010923,
      "eval_loss": 0.28045815229415894,
      "eval_runtime": 53.8781,
      "eval_samples_per_second": 1.726,
      "eval_steps_per_second": 1.726,
      "step": 1000
    },
    {
      "epoch": 0.3276897870016384,
      "grad_norm": 0.80877685546875,
      "learning_rate": 4.3693562481794354e-05,
      "loss": 0.13,
      "step": 1500
    },
    {
      "epoch": 0.3276897870016384,
      "eval_loss": 0.24824748933315277,
      "eval_runtime": 53.7906,
      "eval_samples_per_second": 1.729,
      "eval_steps_per_second": 1.729,
      "step": 1500
    },
    {
      "epoch": 0.4369197160021846,
      "grad_norm": 0.7288264632225037,
      "learning_rate": 5.8228954267404603e-05,
      "loss": 0.1067,
      "step": 2000
    },
    {
      "epoch": 0.4369197160021846,
      "eval_loss": 0.24914385378360748,
      "eval_runtime": 53.8469,
      "eval_samples_per_second": 1.727,
      "eval_steps_per_second": 1.727,
      "step": 2000
    },
    {
      "epoch": 0.5461496450027308,
      "grad_norm": 1.1100046634674072,
      "learning_rate": 7.27934750946694e-05,
      "loss": 0.1018,
      "step": 2500
    },
    {
      "epoch": 0.5461496450027308,
      "eval_loss": 0.27210181951522827,
      "eval_runtime": 53.5513,
      "eval_samples_per_second": 1.737,
      "eval_steps_per_second": 1.737,
      "step": 2500
    },
    {
      "epoch": 0.6553795740032768,
      "grad_norm": 0.8480390310287476,
      "learning_rate": 8.732886688027965e-05,
      "loss": 0.0976,
      "step": 3000
    },
    {
      "epoch": 0.6553795740032768,
      "eval_loss": 0.2787032425403595,
      "eval_runtime": 51.6926,
      "eval_samples_per_second": 1.799,
      "eval_steps_per_second": 1.799,
      "step": 3000
    },
    {
      "epoch": 0.764609503003823,
      "grad_norm": 0.8974778056144714,
      "learning_rate": 0.00010189338770754442,
      "loss": 0.0948,
      "step": 3500
    },
    {
      "epoch": 0.764609503003823,
      "eval_loss": 0.26553070545196533,
      "eval_runtime": 52.0101,
      "eval_samples_per_second": 1.788,
      "eval_steps_per_second": 1.788,
      "step": 3500
    },
    {
      "epoch": 0.8738394320043692,
      "grad_norm": 0.4318696856498718,
      "learning_rate": 0.00011645790853480921,
      "loss": 0.0941,
      "step": 4000
    },
    {
      "epoch": 0.8738394320043692,
      "eval_loss": 0.2772391140460968,
      "eval_runtime": 51.821,
      "eval_samples_per_second": 1.795,
      "eval_steps_per_second": 1.795,
      "step": 4000
    },
    {
      "epoch": 0.9830693610049154,
      "grad_norm": 0.22688086330890656,
      "learning_rate": 0.00013102242936207398,
      "loss": 0.0916,
      "step": 4500
    },
    {
      "epoch": 0.9830693610049154,
      "eval_loss": 0.2516581118106842,
      "eval_runtime": 51.8197,
      "eval_samples_per_second": 1.795,
      "eval_steps_per_second": 1.795,
      "step": 4500
    },
    {
      "epoch": 1.0922992900054616,
      "grad_norm": 0.053704842925071716,
      "learning_rate": 0.0001455869501893388,
      "loss": 0.084,
      "step": 5000
    },
    {
      "epoch": 1.0922992900054616,
      "eval_loss": 0.24150623381137848,
      "eval_runtime": 50.6724,
      "eval_samples_per_second": 1.835,
      "eval_steps_per_second": 1.835,
      "step": 5000
    },
    {
      "epoch": 1.2015292190060076,
      "grad_norm": 0.30040594935417175,
      "learning_rate": 0.00016015147101660357,
      "loss": 0.0882,
      "step": 5500
    },
    {
      "epoch": 1.2015292190060076,
      "eval_loss": 0.24166980385780334,
      "eval_runtime": 50.7195,
      "eval_samples_per_second": 1.834,
      "eval_steps_per_second": 1.834,
      "step": 5500
    },
    {
      "epoch": 1.3107591480065537,
      "grad_norm": 0.21068286895751953,
      "learning_rate": 0.00017471599184386836,
      "loss": 0.0868,
      "step": 6000
    },
    {
      "epoch": 1.3107591480065537,
      "eval_loss": 0.2663758397102356,
      "eval_runtime": 50.6208,
      "eval_samples_per_second": 1.837,
      "eval_steps_per_second": 1.837,
      "step": 6000
    },
    {
      "epoch": 1.4199890770071,
      "grad_norm": 0.1310437172651291,
      "learning_rate": 0.00018928051267113314,
      "loss": 0.087,
      "step": 6500
    },
    {
      "epoch": 1.4199890770071,
      "eval_loss": 0.2654285728931427,
      "eval_runtime": 50.5893,
      "eval_samples_per_second": 1.838,
      "eval_steps_per_second": 1.838,
      "step": 6500
    },
    {
      "epoch": 1.529219006007646,
      "grad_norm": 0.1630202829837799,
      "learning_rate": 0.00019988197347556582,
      "loss": 0.0841,
      "step": 7000
    },
    {
      "epoch": 1.529219006007646,
      "eval_loss": 0.25736120343208313,
      "eval_runtime": 50.5692,
      "eval_samples_per_second": 1.839,
      "eval_steps_per_second": 1.839,
      "step": 7000
    },
    {
      "epoch": 1.6385581649371928,
      "grad_norm": 0.07697170972824097,
      "learning_rate": 0.00019943149055787806,
      "loss": 0.0807,
      "step": 7500
    },
    {
      "epoch": 1.6385581649371928,
      "eval_loss": 0.265057772397995,
      "eval_runtime": 50.3751,
      "eval_samples_per_second": 1.846,
      "eval_steps_per_second": 1.846,
      "step": 7500
    },
    {
      "epoch": 1.7477880939377388,
      "grad_norm": 0.10595889389514923,
      "learning_rate": 0.0001989810076401903,
      "loss": 0.0782,
      "step": 8000
    },
    {
      "epoch": 1.7477880939377388,
      "eval_loss": 0.2732037305831909,
      "eval_runtime": 50.5346,
      "eval_samples_per_second": 1.84,
      "eval_steps_per_second": 1.84,
      "step": 8000
    },
    {
      "epoch": 1.857018022938285,
      "grad_norm": 0.13792632520198822,
      "learning_rate": 0.00019853052472250252,
      "loss": 0.0757,
      "step": 8500
    },
    {
      "epoch": 1.857018022938285,
      "eval_loss": 0.25934740900993347,
      "eval_runtime": 50.3603,
      "eval_samples_per_second": 1.847,
      "eval_steps_per_second": 1.847,
      "step": 8500
    },
    {
      "epoch": 1.97,
      "grad_norm": 0.1341710239648819,
      "learning_rate": 0.00019808004180481476,
      "loss": 0.0764,
      "step": 9000
    },
    {
      "epoch": 1.97,
      "eval_loss": 0.25228968262672424,
      "eval_runtime": 47.204,
      "eval_samples_per_second": 1.97,
      "eval_steps_per_second": 1.97,
      "step": 9000
    },
    {
      "epoch": 2.16,
      "grad_norm": 0.011377187445759773,
      "learning_rate": 0.00019726869847708111,
      "loss": 0.0743,
      "step": 9500
    },
    {
      "epoch": 2.16,
      "eval_loss": 0.1291830837726593,
      "eval_runtime": 193.506,
      "eval_samples_per_second": 2.393,
      "eval_steps_per_second": 2.393,
      "step": 9500
    },
    {
      "epoch": 2.184817039868924,
      "grad_norm": 0.39175206422805786,
      "learning_rate": 0.00019717907596943925,
      "loss": 0.0717,
      "step": 10000
    },
    {
      "epoch": 2.184817039868924,
      "eval_loss": 0.2626956105232239,
      "eval_runtime": 51.1714,
      "eval_samples_per_second": 1.817,
      "eval_steps_per_second": 1.817,
      "step": 10000
    },
    {
      "epoch": 2.2940469688694702,
      "grad_norm": 0.17580239474773407,
      "learning_rate": 0.0001967285930517515,
      "loss": 0.0703,
      "step": 10500
    },
    {
      "epoch": 2.2940469688694702,
      "eval_loss": 0.28000086545944214,
      "eval_runtime": 51.2031,
      "eval_samples_per_second": 1.816,
      "eval_steps_per_second": 1.816,
      "step": 10500
    },
    {
      "epoch": 2.4032768978700165,
      "grad_norm": 0.27358973026275635,
      "learning_rate": 0.00019627811013406371,
      "loss": 0.0707,
      "step": 11000
    },
    {
      "epoch": 2.4032768978700165,
      "eval_loss": 0.27782928943634033,
      "eval_runtime": 50.7155,
      "eval_samples_per_second": 1.834,
      "eval_steps_per_second": 1.834,
      "step": 11000
    }
  ],
  "logging_steps": 500,
  "max_steps": 228850,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "total_flos": 2.298922015943301e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}