sedrickkeh commited on
Commit
12e4f15
·
verified ·
1 Parent(s): f1cb172

End of training

Browse files
README.md CHANGED
@@ -4,6 +4,7 @@ license: llama3.1
4
  base_model: meta-llama/Meta-Llama-3.1-8B
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: tulu-3-sft-mixture
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # tulu-3-sft-mixture
17
 
18
- This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 0.6146
21
 
 
4
  base_model: meta-llama/Meta-Llama-3.1-8B
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: tulu-3-sft-mixture
 
16
 
17
  # tulu-3-sft-mixture
18
 
19
+ This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on the allenai/tulu-3-sft-mixture dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.6146
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_loss": 0.6146459579467773,
4
+ "eval_runtime": 155.7964,
5
+ "eval_samples_per_second": 99.232,
6
+ "eval_steps_per_second": 0.392,
7
+ "total_flos": 2883576618024960.0,
8
+ "train_loss": 0.5825082723410426,
9
+ "train_runtime": 26219.2762,
10
+ "train_samples_per_second": 33.609,
11
+ "train_steps_per_second": 0.066
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_loss": 0.6146459579467773,
4
+ "eval_runtime": 155.7964,
5
+ "eval_samples_per_second": 99.232,
6
+ "eval_steps_per_second": 0.392
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "total_flos": 2883576618024960.0,
4
+ "train_loss": 0.5825082723410426,
5
+ "train_runtime": 26219.2762,
6
+ "train_samples_per_second": 33.609,
7
+ "train_steps_per_second": 0.066
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1270 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0,
5
+ "eval_steps": 500,
6
+ "global_step": 1722,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.017421602787456445,
13
+ "grad_norm": 1.110547889420809,
14
+ "learning_rate": 5e-06,
15
+ "loss": 0.8085,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.03484320557491289,
20
+ "grad_norm": 0.9516650393814576,
21
+ "learning_rate": 5e-06,
22
+ "loss": 0.7273,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.05226480836236934,
27
+ "grad_norm": 0.8745560758252257,
28
+ "learning_rate": 5e-06,
29
+ "loss": 0.7151,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.06968641114982578,
34
+ "grad_norm": 0.6458692027569163,
35
+ "learning_rate": 5e-06,
36
+ "loss": 0.7042,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.08710801393728224,
41
+ "grad_norm": 0.6419702080069286,
42
+ "learning_rate": 5e-06,
43
+ "loss": 0.6797,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.10452961672473868,
48
+ "grad_norm": 1.3167787941292113,
49
+ "learning_rate": 5e-06,
50
+ "loss": 0.6939,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.12195121951219512,
55
+ "grad_norm": 0.5165069959452926,
56
+ "learning_rate": 5e-06,
57
+ "loss": 0.6731,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.13937282229965156,
62
+ "grad_norm": 0.6494172258411438,
63
+ "learning_rate": 5e-06,
64
+ "loss": 0.6653,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.156794425087108,
69
+ "grad_norm": 0.8433361988376273,
70
+ "learning_rate": 5e-06,
71
+ "loss": 0.6834,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.17421602787456447,
76
+ "grad_norm": 0.7333617572332245,
77
+ "learning_rate": 5e-06,
78
+ "loss": 0.674,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.1916376306620209,
83
+ "grad_norm": 0.46328020438675316,
84
+ "learning_rate": 5e-06,
85
+ "loss": 0.6705,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.20905923344947736,
90
+ "grad_norm": 0.5375878274323563,
91
+ "learning_rate": 5e-06,
92
+ "loss": 0.6546,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.2264808362369338,
97
+ "grad_norm": 0.6043348556726694,
98
+ "learning_rate": 5e-06,
99
+ "loss": 0.6506,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.24390243902439024,
104
+ "grad_norm": 0.4508554439777698,
105
+ "learning_rate": 5e-06,
106
+ "loss": 0.6536,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.2613240418118467,
111
+ "grad_norm": 1.1359138470696808,
112
+ "learning_rate": 5e-06,
113
+ "loss": 0.638,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.2787456445993031,
118
+ "grad_norm": 0.5757321381081676,
119
+ "learning_rate": 5e-06,
120
+ "loss": 0.6492,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.2961672473867596,
125
+ "grad_norm": 0.49845234216314765,
126
+ "learning_rate": 5e-06,
127
+ "loss": 0.6481,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.313588850174216,
132
+ "grad_norm": 0.4885975758468272,
133
+ "learning_rate": 5e-06,
134
+ "loss": 0.6445,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.3310104529616725,
139
+ "grad_norm": 0.46363877509236,
140
+ "learning_rate": 5e-06,
141
+ "loss": 0.6437,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.34843205574912894,
146
+ "grad_norm": 0.4685718300832952,
147
+ "learning_rate": 5e-06,
148
+ "loss": 0.6386,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.36585365853658536,
153
+ "grad_norm": 0.5605596050023267,
154
+ "learning_rate": 5e-06,
155
+ "loss": 0.6507,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.3832752613240418,
160
+ "grad_norm": 0.6403424736477891,
161
+ "learning_rate": 5e-06,
162
+ "loss": 0.6418,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.40069686411149824,
167
+ "grad_norm": 0.5838158484945601,
168
+ "learning_rate": 5e-06,
169
+ "loss": 0.6493,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.4181184668989547,
174
+ "grad_norm": 0.45330751179903367,
175
+ "learning_rate": 5e-06,
176
+ "loss": 0.6498,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.4355400696864111,
181
+ "grad_norm": 0.47780598447174244,
182
+ "learning_rate": 5e-06,
183
+ "loss": 0.6359,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.4529616724738676,
188
+ "grad_norm": 0.4754778849361075,
189
+ "learning_rate": 5e-06,
190
+ "loss": 0.6402,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.47038327526132406,
195
+ "grad_norm": 0.9267353139931963,
196
+ "learning_rate": 5e-06,
197
+ "loss": 0.6453,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.4878048780487805,
202
+ "grad_norm": 1.09959569850648,
203
+ "learning_rate": 5e-06,
204
+ "loss": 0.6381,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.5052264808362369,
209
+ "grad_norm": 0.46907276389959596,
210
+ "learning_rate": 5e-06,
211
+ "loss": 0.6142,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.5226480836236934,
216
+ "grad_norm": 0.9009500821986945,
217
+ "learning_rate": 5e-06,
218
+ "loss": 0.6349,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.5400696864111498,
223
+ "grad_norm": 0.8732778985630532,
224
+ "learning_rate": 5e-06,
225
+ "loss": 0.633,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.5574912891986062,
230
+ "grad_norm": 0.5330855012377655,
231
+ "learning_rate": 5e-06,
232
+ "loss": 0.6292,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.5749128919860628,
237
+ "grad_norm": 0.5021412884291475,
238
+ "learning_rate": 5e-06,
239
+ "loss": 0.6253,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.5923344947735192,
244
+ "grad_norm": 0.5447288950929059,
245
+ "learning_rate": 5e-06,
246
+ "loss": 0.6308,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.6097560975609756,
251
+ "grad_norm": 0.7644080552686335,
252
+ "learning_rate": 5e-06,
253
+ "loss": 0.6267,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.627177700348432,
258
+ "grad_norm": 0.4607717863440372,
259
+ "learning_rate": 5e-06,
260
+ "loss": 0.637,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.6445993031358885,
265
+ "grad_norm": 0.6059522443076883,
266
+ "learning_rate": 5e-06,
267
+ "loss": 0.6276,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.662020905923345,
272
+ "grad_norm": 0.6368838367327173,
273
+ "learning_rate": 5e-06,
274
+ "loss": 0.6227,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.6794425087108014,
279
+ "grad_norm": 0.4571682914981839,
280
+ "learning_rate": 5e-06,
281
+ "loss": 0.6204,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.6968641114982579,
286
+ "grad_norm": 0.7704076665510348,
287
+ "learning_rate": 5e-06,
288
+ "loss": 0.6332,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.7142857142857143,
293
+ "grad_norm": 0.48016934744783224,
294
+ "learning_rate": 5e-06,
295
+ "loss": 0.6317,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.7317073170731707,
300
+ "grad_norm": 0.4659861774424128,
301
+ "learning_rate": 5e-06,
302
+ "loss": 0.614,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.7491289198606271,
307
+ "grad_norm": 0.5709628311962424,
308
+ "learning_rate": 5e-06,
309
+ "loss": 0.617,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.7665505226480837,
314
+ "grad_norm": 0.9275541597160887,
315
+ "learning_rate": 5e-06,
316
+ "loss": 0.6259,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 0.7839721254355401,
321
+ "grad_norm": 0.4590863644330183,
322
+ "learning_rate": 5e-06,
323
+ "loss": 0.6432,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 0.8013937282229965,
328
+ "grad_norm": 0.44311113231679206,
329
+ "learning_rate": 5e-06,
330
+ "loss": 0.6325,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 0.818815331010453,
335
+ "grad_norm": 0.42872958673136763,
336
+ "learning_rate": 5e-06,
337
+ "loss": 0.6254,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 0.8362369337979094,
342
+ "grad_norm": 0.6968494949424339,
343
+ "learning_rate": 5e-06,
344
+ "loss": 0.6282,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 0.8536585365853658,
349
+ "grad_norm": 0.475991569929859,
350
+ "learning_rate": 5e-06,
351
+ "loss": 0.6104,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 0.8710801393728222,
356
+ "grad_norm": 0.4036764356634414,
357
+ "learning_rate": 5e-06,
358
+ "loss": 0.6192,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 0.8885017421602788,
363
+ "grad_norm": 0.4800314428435892,
364
+ "learning_rate": 5e-06,
365
+ "loss": 0.6226,
366
+ "step": 510
367
+ },
368
+ {
369
+ "epoch": 0.9059233449477352,
370
+ "grad_norm": 0.4448196930678713,
371
+ "learning_rate": 5e-06,
372
+ "loss": 0.6141,
373
+ "step": 520
374
+ },
375
+ {
376
+ "epoch": 0.9233449477351916,
377
+ "grad_norm": 0.43823469299056167,
378
+ "learning_rate": 5e-06,
379
+ "loss": 0.6137,
380
+ "step": 530
381
+ },
382
+ {
383
+ "epoch": 0.9407665505226481,
384
+ "grad_norm": 0.46401352050703015,
385
+ "learning_rate": 5e-06,
386
+ "loss": 0.6155,
387
+ "step": 540
388
+ },
389
+ {
390
+ "epoch": 0.9581881533101045,
391
+ "grad_norm": 0.4741335427712111,
392
+ "learning_rate": 5e-06,
393
+ "loss": 0.6239,
394
+ "step": 550
395
+ },
396
+ {
397
+ "epoch": 0.975609756097561,
398
+ "grad_norm": 0.5404449413086233,
399
+ "learning_rate": 5e-06,
400
+ "loss": 0.6228,
401
+ "step": 560
402
+ },
403
+ {
404
+ "epoch": 0.9930313588850174,
405
+ "grad_norm": 0.4865700073464584,
406
+ "learning_rate": 5e-06,
407
+ "loss": 0.6057,
408
+ "step": 570
409
+ },
410
+ {
411
+ "epoch": 1.0,
412
+ "eval_loss": 0.6242489218711853,
413
+ "eval_runtime": 156.3164,
414
+ "eval_samples_per_second": 98.902,
415
+ "eval_steps_per_second": 0.39,
416
+ "step": 574
417
+ },
418
+ {
419
+ "epoch": 1.0104529616724738,
420
+ "grad_norm": 0.49999642055125576,
421
+ "learning_rate": 5e-06,
422
+ "loss": 0.5959,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 1.0278745644599303,
427
+ "grad_norm": 0.45102982348932047,
428
+ "learning_rate": 5e-06,
429
+ "loss": 0.5783,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 1.0452961672473868,
434
+ "grad_norm": 0.4675746175963686,
435
+ "learning_rate": 5e-06,
436
+ "loss": 0.5907,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 1.0627177700348431,
441
+ "grad_norm": 0.4332204857821093,
442
+ "learning_rate": 5e-06,
443
+ "loss": 0.5721,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 1.0801393728222997,
448
+ "grad_norm": 0.46797084979476816,
449
+ "learning_rate": 5e-06,
450
+ "loss": 0.5728,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 1.0975609756097562,
455
+ "grad_norm": 0.45462105008725134,
456
+ "learning_rate": 5e-06,
457
+ "loss": 0.5766,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 1.1149825783972125,
462
+ "grad_norm": 0.45583632727945284,
463
+ "learning_rate": 5e-06,
464
+ "loss": 0.5684,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 1.132404181184669,
469
+ "grad_norm": 0.6045690088290018,
470
+ "learning_rate": 5e-06,
471
+ "loss": 0.5839,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 1.1498257839721253,
476
+ "grad_norm": 0.4772407436839459,
477
+ "learning_rate": 5e-06,
478
+ "loss": 0.5745,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 1.1672473867595818,
483
+ "grad_norm": 0.7134135897555454,
484
+ "learning_rate": 5e-06,
485
+ "loss": 0.5763,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 1.1846689895470384,
490
+ "grad_norm": 0.5626507275472277,
491
+ "learning_rate": 5e-06,
492
+ "loss": 0.5699,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 1.202090592334495,
497
+ "grad_norm": 0.4587849239706485,
498
+ "learning_rate": 5e-06,
499
+ "loss": 0.5975,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 1.2195121951219512,
504
+ "grad_norm": 0.45602175911022613,
505
+ "learning_rate": 5e-06,
506
+ "loss": 0.5727,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 1.2369337979094077,
511
+ "grad_norm": 0.4760989235025255,
512
+ "learning_rate": 5e-06,
513
+ "loss": 0.5743,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 1.254355400696864,
518
+ "grad_norm": 0.57636457602915,
519
+ "learning_rate": 5e-06,
520
+ "loss": 0.5723,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 1.2717770034843205,
525
+ "grad_norm": 0.7070463930124464,
526
+ "learning_rate": 5e-06,
527
+ "loss": 0.5734,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 1.289198606271777,
532
+ "grad_norm": 0.451570248773914,
533
+ "learning_rate": 5e-06,
534
+ "loss": 0.5654,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 1.3066202090592334,
539
+ "grad_norm": 0.4634040338505886,
540
+ "learning_rate": 5e-06,
541
+ "loss": 0.5734,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 1.32404181184669,
546
+ "grad_norm": 0.461184295216135,
547
+ "learning_rate": 5e-06,
548
+ "loss": 0.5659,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 1.3414634146341464,
553
+ "grad_norm": 0.5502438318822999,
554
+ "learning_rate": 5e-06,
555
+ "loss": 0.5638,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 1.3588850174216027,
560
+ "grad_norm": 0.40334764475120544,
561
+ "learning_rate": 5e-06,
562
+ "loss": 0.5741,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 1.3763066202090593,
567
+ "grad_norm": 0.46619326588491156,
568
+ "learning_rate": 5e-06,
569
+ "loss": 0.5816,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 1.3937282229965158,
574
+ "grad_norm": 0.43450365905550786,
575
+ "learning_rate": 5e-06,
576
+ "loss": 0.5879,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 1.411149825783972,
581
+ "grad_norm": 0.4491332694126357,
582
+ "learning_rate": 5e-06,
583
+ "loss": 0.5834,
584
+ "step": 810
585
+ },
586
+ {
587
+ "epoch": 1.4285714285714286,
588
+ "grad_norm": 0.4492632982117168,
589
+ "learning_rate": 5e-06,
590
+ "loss": 0.5704,
591
+ "step": 820
592
+ },
593
+ {
594
+ "epoch": 1.445993031358885,
595
+ "grad_norm": 0.48204516908878015,
596
+ "learning_rate": 5e-06,
597
+ "loss": 0.5726,
598
+ "step": 830
599
+ },
600
+ {
601
+ "epoch": 1.4634146341463414,
602
+ "grad_norm": 0.4451434847610245,
603
+ "learning_rate": 5e-06,
604
+ "loss": 0.5725,
605
+ "step": 840
606
+ },
607
+ {
608
+ "epoch": 1.480836236933798,
609
+ "grad_norm": 0.4817077921497448,
610
+ "learning_rate": 5e-06,
611
+ "loss": 0.578,
612
+ "step": 850
613
+ },
614
+ {
615
+ "epoch": 1.4982578397212545,
616
+ "grad_norm": 0.4114565463440688,
617
+ "learning_rate": 5e-06,
618
+ "loss": 0.5756,
619
+ "step": 860
620
+ },
621
+ {
622
+ "epoch": 1.5156794425087108,
623
+ "grad_norm": 0.4364153090393298,
624
+ "learning_rate": 5e-06,
625
+ "loss": 0.5781,
626
+ "step": 870
627
+ },
628
+ {
629
+ "epoch": 1.533101045296167,
630
+ "grad_norm": 0.4641250735638994,
631
+ "learning_rate": 5e-06,
632
+ "loss": 0.5858,
633
+ "step": 880
634
+ },
635
+ {
636
+ "epoch": 1.5505226480836236,
637
+ "grad_norm": 0.5414804208775645,
638
+ "learning_rate": 5e-06,
639
+ "loss": 0.576,
640
+ "step": 890
641
+ },
642
+ {
643
+ "epoch": 1.5679442508710801,
644
+ "grad_norm": 0.4046063470124716,
645
+ "learning_rate": 5e-06,
646
+ "loss": 0.5764,
647
+ "step": 900
648
+ },
649
+ {
650
+ "epoch": 1.5853658536585367,
651
+ "grad_norm": 0.5021694200790203,
652
+ "learning_rate": 5e-06,
653
+ "loss": 0.5832,
654
+ "step": 910
655
+ },
656
+ {
657
+ "epoch": 1.6027874564459932,
658
+ "grad_norm": 0.44359324558365465,
659
+ "learning_rate": 5e-06,
660
+ "loss": 0.5702,
661
+ "step": 920
662
+ },
663
+ {
664
+ "epoch": 1.6202090592334495,
665
+ "grad_norm": 0.4590082845753106,
666
+ "learning_rate": 5e-06,
667
+ "loss": 0.5702,
668
+ "step": 930
669
+ },
670
+ {
671
+ "epoch": 1.6376306620209058,
672
+ "grad_norm": 0.42340672053613443,
673
+ "learning_rate": 5e-06,
674
+ "loss": 0.5788,
675
+ "step": 940
676
+ },
677
+ {
678
+ "epoch": 1.6550522648083623,
679
+ "grad_norm": 0.4144352252936342,
680
+ "learning_rate": 5e-06,
681
+ "loss": 0.5766,
682
+ "step": 950
683
+ },
684
+ {
685
+ "epoch": 1.6724738675958188,
686
+ "grad_norm": 0.4171029941084573,
687
+ "learning_rate": 5e-06,
688
+ "loss": 0.5695,
689
+ "step": 960
690
+ },
691
+ {
692
+ "epoch": 1.6898954703832754,
693
+ "grad_norm": 0.43911657202438914,
694
+ "learning_rate": 5e-06,
695
+ "loss": 0.5661,
696
+ "step": 970
697
+ },
698
+ {
699
+ "epoch": 1.7073170731707317,
700
+ "grad_norm": 0.4114986035010171,
701
+ "learning_rate": 5e-06,
702
+ "loss": 0.571,
703
+ "step": 980
704
+ },
705
+ {
706
+ "epoch": 1.7247386759581882,
707
+ "grad_norm": 0.43109782590499385,
708
+ "learning_rate": 5e-06,
709
+ "loss": 0.5747,
710
+ "step": 990
711
+ },
712
+ {
713
+ "epoch": 1.7421602787456445,
714
+ "grad_norm": 0.4565580100991871,
715
+ "learning_rate": 5e-06,
716
+ "loss": 0.5771,
717
+ "step": 1000
718
+ },
719
+ {
720
+ "epoch": 1.759581881533101,
721
+ "grad_norm": 0.4138828268586911,
722
+ "learning_rate": 5e-06,
723
+ "loss": 0.5629,
724
+ "step": 1010
725
+ },
726
+ {
727
+ "epoch": 1.7770034843205575,
728
+ "grad_norm": 0.4542995316054866,
729
+ "learning_rate": 5e-06,
730
+ "loss": 0.5856,
731
+ "step": 1020
732
+ },
733
+ {
734
+ "epoch": 1.794425087108014,
735
+ "grad_norm": 0.6622604112941458,
736
+ "learning_rate": 5e-06,
737
+ "loss": 0.5679,
738
+ "step": 1030
739
+ },
740
+ {
741
+ "epoch": 1.8118466898954704,
742
+ "grad_norm": 0.4507517659371942,
743
+ "learning_rate": 5e-06,
744
+ "loss": 0.5774,
745
+ "step": 1040
746
+ },
747
+ {
748
+ "epoch": 1.8292682926829267,
749
+ "grad_norm": 0.4309645347763417,
750
+ "learning_rate": 5e-06,
751
+ "loss": 0.5704,
752
+ "step": 1050
753
+ },
754
+ {
755
+ "epoch": 1.8466898954703832,
756
+ "grad_norm": 0.41185890423116756,
757
+ "learning_rate": 5e-06,
758
+ "loss": 0.5734,
759
+ "step": 1060
760
+ },
761
+ {
762
+ "epoch": 1.8641114982578397,
763
+ "grad_norm": 0.4455851719114888,
764
+ "learning_rate": 5e-06,
765
+ "loss": 0.5747,
766
+ "step": 1070
767
+ },
768
+ {
769
+ "epoch": 1.8815331010452963,
770
+ "grad_norm": 0.41686045633860264,
771
+ "learning_rate": 5e-06,
772
+ "loss": 0.5771,
773
+ "step": 1080
774
+ },
775
+ {
776
+ "epoch": 1.8989547038327528,
777
+ "grad_norm": 0.4333938820779194,
778
+ "learning_rate": 5e-06,
779
+ "loss": 0.579,
780
+ "step": 1090
781
+ },
782
+ {
783
+ "epoch": 1.916376306620209,
784
+ "grad_norm": 0.4045207725207787,
785
+ "learning_rate": 5e-06,
786
+ "loss": 0.5758,
787
+ "step": 1100
788
+ },
789
+ {
790
+ "epoch": 1.9337979094076654,
791
+ "grad_norm": 0.4520222740004976,
792
+ "learning_rate": 5e-06,
793
+ "loss": 0.5563,
794
+ "step": 1110
795
+ },
796
+ {
797
+ "epoch": 1.951219512195122,
798
+ "grad_norm": 0.4311997583086096,
799
+ "learning_rate": 5e-06,
800
+ "loss": 0.5681,
801
+ "step": 1120
802
+ },
803
+ {
804
+ "epoch": 1.9686411149825784,
805
+ "grad_norm": 0.4383316797763638,
806
+ "learning_rate": 5e-06,
807
+ "loss": 0.5819,
808
+ "step": 1130
809
+ },
810
+ {
811
+ "epoch": 1.986062717770035,
812
+ "grad_norm": 0.5477336634704406,
813
+ "learning_rate": 5e-06,
814
+ "loss": 0.5711,
815
+ "step": 1140
816
+ },
817
+ {
818
+ "epoch": 2.0,
819
+ "eval_loss": 0.6141585111618042,
820
+ "eval_runtime": 156.1142,
821
+ "eval_samples_per_second": 99.03,
822
+ "eval_steps_per_second": 0.391,
823
+ "step": 1148
824
+ },
825
+ {
826
+ "epoch": 2.0034843205574915,
827
+ "grad_norm": 0.4640944945247334,
828
+ "learning_rate": 5e-06,
829
+ "loss": 0.5571,
830
+ "step": 1150
831
+ },
832
+ {
833
+ "epoch": 2.0209059233449476,
834
+ "grad_norm": 0.45843359238558995,
835
+ "learning_rate": 5e-06,
836
+ "loss": 0.5249,
837
+ "step": 1160
838
+ },
839
+ {
840
+ "epoch": 2.038327526132404,
841
+ "grad_norm": 0.4837328490077125,
842
+ "learning_rate": 5e-06,
843
+ "loss": 0.5282,
844
+ "step": 1170
845
+ },
846
+ {
847
+ "epoch": 2.0557491289198606,
848
+ "grad_norm": 0.44590774801900834,
849
+ "learning_rate": 5e-06,
850
+ "loss": 0.5337,
851
+ "step": 1180
852
+ },
853
+ {
854
+ "epoch": 2.073170731707317,
855
+ "grad_norm": 0.43363076292724995,
856
+ "learning_rate": 5e-06,
857
+ "loss": 0.5374,
858
+ "step": 1190
859
+ },
860
+ {
861
+ "epoch": 2.0905923344947737,
862
+ "grad_norm": 0.4931511449549238,
863
+ "learning_rate": 5e-06,
864
+ "loss": 0.5309,
865
+ "step": 1200
866
+ },
867
+ {
868
+ "epoch": 2.10801393728223,
869
+ "grad_norm": 0.4504251344307032,
870
+ "learning_rate": 5e-06,
871
+ "loss": 0.5271,
872
+ "step": 1210
873
+ },
874
+ {
875
+ "epoch": 2.1254355400696863,
876
+ "grad_norm": 0.49526543085134606,
877
+ "learning_rate": 5e-06,
878
+ "loss": 0.5319,
879
+ "step": 1220
880
+ },
881
+ {
882
+ "epoch": 2.142857142857143,
883
+ "grad_norm": 0.4677844909297682,
884
+ "learning_rate": 5e-06,
885
+ "loss": 0.5306,
886
+ "step": 1230
887
+ },
888
+ {
889
+ "epoch": 2.1602787456445993,
890
+ "grad_norm": 0.47260212803851326,
891
+ "learning_rate": 5e-06,
892
+ "loss": 0.5228,
893
+ "step": 1240
894
+ },
895
+ {
896
+ "epoch": 2.177700348432056,
897
+ "grad_norm": 0.5490671664415253,
898
+ "learning_rate": 5e-06,
899
+ "loss": 0.5276,
900
+ "step": 1250
901
+ },
902
+ {
903
+ "epoch": 2.1951219512195124,
904
+ "grad_norm": 0.4780364714560436,
905
+ "learning_rate": 5e-06,
906
+ "loss": 0.5307,
907
+ "step": 1260
908
+ },
909
+ {
910
+ "epoch": 2.2125435540069684,
911
+ "grad_norm": 0.44095382556774654,
912
+ "learning_rate": 5e-06,
913
+ "loss": 0.5273,
914
+ "step": 1270
915
+ },
916
+ {
917
+ "epoch": 2.229965156794425,
918
+ "grad_norm": 0.471163066518852,
919
+ "learning_rate": 5e-06,
920
+ "loss": 0.5095,
921
+ "step": 1280
922
+ },
923
+ {
924
+ "epoch": 2.2473867595818815,
925
+ "grad_norm": 0.4509784260620687,
926
+ "learning_rate": 5e-06,
927
+ "loss": 0.525,
928
+ "step": 1290
929
+ },
930
+ {
931
+ "epoch": 2.264808362369338,
932
+ "grad_norm": 0.42926159657173546,
933
+ "learning_rate": 5e-06,
934
+ "loss": 0.5254,
935
+ "step": 1300
936
+ },
937
+ {
938
+ "epoch": 2.2822299651567945,
939
+ "grad_norm": 0.47746501662453417,
940
+ "learning_rate": 5e-06,
941
+ "loss": 0.5229,
942
+ "step": 1310
943
+ },
944
+ {
945
+ "epoch": 2.2996515679442506,
946
+ "grad_norm": 0.4747495543682741,
947
+ "learning_rate": 5e-06,
948
+ "loss": 0.5259,
949
+ "step": 1320
950
+ },
951
+ {
952
+ "epoch": 2.317073170731707,
953
+ "grad_norm": 0.5228901131403929,
954
+ "learning_rate": 5e-06,
955
+ "loss": 0.5243,
956
+ "step": 1330
957
+ },
958
+ {
959
+ "epoch": 2.3344947735191637,
960
+ "grad_norm": 0.4834167699517272,
961
+ "learning_rate": 5e-06,
962
+ "loss": 0.5306,
963
+ "step": 1340
964
+ },
965
+ {
966
+ "epoch": 2.35191637630662,
967
+ "grad_norm": 0.45707310674276885,
968
+ "learning_rate": 5e-06,
969
+ "loss": 0.5276,
970
+ "step": 1350
971
+ },
972
+ {
973
+ "epoch": 2.3693379790940767,
974
+ "grad_norm": 0.47685998083437814,
975
+ "learning_rate": 5e-06,
976
+ "loss": 0.5312,
977
+ "step": 1360
978
+ },
979
+ {
980
+ "epoch": 2.3867595818815333,
981
+ "grad_norm": 0.47877017672465005,
982
+ "learning_rate": 5e-06,
983
+ "loss": 0.5274,
984
+ "step": 1370
985
+ },
986
+ {
987
+ "epoch": 2.40418118466899,
988
+ "grad_norm": 0.5258496984168791,
989
+ "learning_rate": 5e-06,
990
+ "loss": 0.5226,
991
+ "step": 1380
992
+ },
993
+ {
994
+ "epoch": 2.421602787456446,
995
+ "grad_norm": 0.47709678060118926,
996
+ "learning_rate": 5e-06,
997
+ "loss": 0.5189,
998
+ "step": 1390
999
+ },
1000
+ {
1001
+ "epoch": 2.4390243902439024,
1002
+ "grad_norm": 0.581254404109644,
1003
+ "learning_rate": 5e-06,
1004
+ "loss": 0.54,
1005
+ "step": 1400
1006
+ },
1007
+ {
1008
+ "epoch": 2.456445993031359,
1009
+ "grad_norm": 0.43351552520462094,
1010
+ "learning_rate": 5e-06,
1011
+ "loss": 0.5345,
1012
+ "step": 1410
1013
+ },
1014
+ {
1015
+ "epoch": 2.4738675958188154,
1016
+ "grad_norm": 0.4485068008247544,
1017
+ "learning_rate": 5e-06,
1018
+ "loss": 0.5361,
1019
+ "step": 1420
1020
+ },
1021
+ {
1022
+ "epoch": 2.491289198606272,
1023
+ "grad_norm": 0.4729416207097551,
1024
+ "learning_rate": 5e-06,
1025
+ "loss": 0.5351,
1026
+ "step": 1430
1027
+ },
1028
+ {
1029
+ "epoch": 2.508710801393728,
1030
+ "grad_norm": 0.44072110157252653,
1031
+ "learning_rate": 5e-06,
1032
+ "loss": 0.5294,
1033
+ "step": 1440
1034
+ },
1035
+ {
1036
+ "epoch": 2.5261324041811846,
1037
+ "grad_norm": 0.5219007442544961,
1038
+ "learning_rate": 5e-06,
1039
+ "loss": 0.5238,
1040
+ "step": 1450
1041
+ },
1042
+ {
1043
+ "epoch": 2.543554006968641,
1044
+ "grad_norm": 0.4809656167510028,
1045
+ "learning_rate": 5e-06,
1046
+ "loss": 0.5288,
1047
+ "step": 1460
1048
+ },
1049
+ {
1050
+ "epoch": 2.5609756097560976,
1051
+ "grad_norm": 0.4531184946595288,
1052
+ "learning_rate": 5e-06,
1053
+ "loss": 0.536,
1054
+ "step": 1470
1055
+ },
1056
+ {
1057
+ "epoch": 2.578397212543554,
1058
+ "grad_norm": 0.48164003860440047,
1059
+ "learning_rate": 5e-06,
1060
+ "loss": 0.5324,
1061
+ "step": 1480
1062
+ },
1063
+ {
1064
+ "epoch": 2.59581881533101,
1065
+ "grad_norm": 0.4758292139458944,
1066
+ "learning_rate": 5e-06,
1067
+ "loss": 0.5133,
1068
+ "step": 1490
1069
+ },
1070
+ {
1071
+ "epoch": 2.6132404181184667,
1072
+ "grad_norm": 0.4634797349634793,
1073
+ "learning_rate": 5e-06,
1074
+ "loss": 0.5295,
1075
+ "step": 1500
1076
+ },
1077
+ {
1078
+ "epoch": 2.6306620209059233,
1079
+ "grad_norm": 0.47375952821313455,
1080
+ "learning_rate": 5e-06,
1081
+ "loss": 0.5273,
1082
+ "step": 1510
1083
+ },
1084
+ {
1085
+ "epoch": 2.64808362369338,
1086
+ "grad_norm": 0.4441843444201693,
1087
+ "learning_rate": 5e-06,
1088
+ "loss": 0.5334,
1089
+ "step": 1520
1090
+ },
1091
+ {
1092
+ "epoch": 2.6655052264808363,
1093
+ "grad_norm": 0.42211766638465,
1094
+ "learning_rate": 5e-06,
1095
+ "loss": 0.5138,
1096
+ "step": 1530
1097
+ },
1098
+ {
1099
+ "epoch": 2.682926829268293,
1100
+ "grad_norm": 0.48817131427198845,
1101
+ "learning_rate": 5e-06,
1102
+ "loss": 0.5134,
1103
+ "step": 1540
1104
+ },
1105
+ {
1106
+ "epoch": 2.7003484320557494,
1107
+ "grad_norm": 0.45942488735176895,
1108
+ "learning_rate": 5e-06,
1109
+ "loss": 0.5361,
1110
+ "step": 1550
1111
+ },
1112
+ {
1113
+ "epoch": 2.7177700348432055,
1114
+ "grad_norm": 0.4854175238598141,
1115
+ "learning_rate": 5e-06,
1116
+ "loss": 0.5394,
1117
+ "step": 1560
1118
+ },
1119
+ {
1120
+ "epoch": 2.735191637630662,
1121
+ "grad_norm": 0.4604289027255158,
1122
+ "learning_rate": 5e-06,
1123
+ "loss": 0.5187,
1124
+ "step": 1570
1125
+ },
1126
+ {
1127
+ "epoch": 2.7526132404181185,
1128
+ "grad_norm": 0.4936790258884402,
1129
+ "learning_rate": 5e-06,
1130
+ "loss": 0.5246,
1131
+ "step": 1580
1132
+ },
1133
+ {
1134
+ "epoch": 2.770034843205575,
1135
+ "grad_norm": 0.541536408959734,
1136
+ "learning_rate": 5e-06,
1137
+ "loss": 0.5278,
1138
+ "step": 1590
1139
+ },
1140
+ {
1141
+ "epoch": 2.7874564459930316,
1142
+ "grad_norm": 0.484918416204369,
1143
+ "learning_rate": 5e-06,
1144
+ "loss": 0.5382,
1145
+ "step": 1600
1146
+ },
1147
+ {
1148
+ "epoch": 2.8048780487804876,
1149
+ "grad_norm": 0.4885761374101617,
1150
+ "learning_rate": 5e-06,
1151
+ "loss": 0.533,
1152
+ "step": 1610
1153
+ },
1154
+ {
1155
+ "epoch": 2.822299651567944,
1156
+ "grad_norm": 0.4349789229036019,
1157
+ "learning_rate": 5e-06,
1158
+ "loss": 0.5388,
1159
+ "step": 1620
1160
+ },
1161
+ {
1162
+ "epoch": 2.8397212543554007,
1163
+ "grad_norm": 0.4314276877403793,
1164
+ "learning_rate": 5e-06,
1165
+ "loss": 0.5391,
1166
+ "step": 1630
1167
+ },
1168
+ {
1169
+ "epoch": 2.857142857142857,
1170
+ "grad_norm": 0.43396532487130346,
1171
+ "learning_rate": 5e-06,
1172
+ "loss": 0.5302,
1173
+ "step": 1640
1174
+ },
1175
+ {
1176
+ "epoch": 2.8745644599303137,
1177
+ "grad_norm": 0.43625746208041316,
1178
+ "learning_rate": 5e-06,
1179
+ "loss": 0.5275,
1180
+ "step": 1650
1181
+ },
1182
+ {
1183
+ "epoch": 2.89198606271777,
1184
+ "grad_norm": 0.4646206916137346,
1185
+ "learning_rate": 5e-06,
1186
+ "loss": 0.5209,
1187
+ "step": 1660
1188
+ },
1189
+ {
1190
+ "epoch": 2.9094076655052263,
1191
+ "grad_norm": 0.4840563025343255,
1192
+ "learning_rate": 5e-06,
1193
+ "loss": 0.5246,
1194
+ "step": 1670
1195
+ },
1196
+ {
1197
+ "epoch": 2.926829268292683,
1198
+ "grad_norm": 0.45066810285940856,
1199
+ "learning_rate": 5e-06,
1200
+ "loss": 0.5318,
1201
+ "step": 1680
1202
+ },
1203
+ {
1204
+ "epoch": 2.9442508710801394,
1205
+ "grad_norm": 0.46846871284532293,
1206
+ "learning_rate": 5e-06,
1207
+ "loss": 0.5324,
1208
+ "step": 1690
1209
+ },
1210
+ {
1211
+ "epoch": 2.961672473867596,
1212
+ "grad_norm": 0.5421178557930114,
1213
+ "learning_rate": 5e-06,
1214
+ "loss": 0.5317,
1215
+ "step": 1700
1216
+ },
1217
+ {
1218
+ "epoch": 2.979094076655052,
1219
+ "grad_norm": 0.44573049149554844,
1220
+ "learning_rate": 5e-06,
1221
+ "loss": 0.5284,
1222
+ "step": 1710
1223
+ },
1224
+ {
1225
+ "epoch": 2.996515679442509,
1226
+ "grad_norm": 0.4720545406402456,
1227
+ "learning_rate": 5e-06,
1228
+ "loss": 0.5286,
1229
+ "step": 1720
1230
+ },
1231
+ {
1232
+ "epoch": 3.0,
1233
+ "eval_loss": 0.6146459579467773,
1234
+ "eval_runtime": 156.0276,
1235
+ "eval_samples_per_second": 99.085,
1236
+ "eval_steps_per_second": 0.391,
1237
+ "step": 1722
1238
+ },
1239
+ {
1240
+ "epoch": 3.0,
1241
+ "step": 1722,
1242
+ "total_flos": 2883576618024960.0,
1243
+ "train_loss": 0.5825082723410426,
1244
+ "train_runtime": 26219.2762,
1245
+ "train_samples_per_second": 33.609,
1246
+ "train_steps_per_second": 0.066
1247
+ }
1248
+ ],
1249
+ "logging_steps": 10,
1250
+ "max_steps": 1722,
1251
+ "num_input_tokens_seen": 0,
1252
+ "num_train_epochs": 3,
1253
+ "save_steps": 500,
1254
+ "stateful_callbacks": {
1255
+ "TrainerControl": {
1256
+ "args": {
1257
+ "should_epoch_stop": false,
1258
+ "should_evaluate": false,
1259
+ "should_log": false,
1260
+ "should_save": true,
1261
+ "should_training_stop": true
1262
+ },
1263
+ "attributes": {}
1264
+ }
1265
+ },
1266
+ "total_flos": 2883576618024960.0,
1267
+ "train_batch_size": 8,
1268
+ "trial_name": null,
1269
+ "trial_params": null
1270
+ }
training_eval_loss.png ADDED
training_loss.png ADDED