sedrickkeh commited on
Commit
033a11f
·
verified ·
1 Parent(s): 5fb0055

End of training

Browse files
Files changed (5) hide show
  1. README.md +2 -1
  2. all_results.json +8 -0
  3. train_results.json +8 -0
  4. trainer_state.json +630 -0
  5. training_loss.png +0 -0
README.md CHANGED
@@ -4,6 +4,7 @@ license: apache-2.0
4
  base_model: Qwen/Qwen2.5-7B-Instruct
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: AM-DeepSeek-R1-Distilled-1.4M-am_0.9M_sample100k
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # AM-DeepSeek-R1-Distilled-1.4M-am_0.9M_sample100k
17
 
18
- This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on an unknown dataset.
19
 
20
  ## Model description
21
 
 
4
  base_model: Qwen/Qwen2.5-7B-Instruct
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: AM-DeepSeek-R1-Distilled-1.4M-am_0.9M_sample100k
 
16
 
17
  # AM-DeepSeek-R1-Distilled-1.4M-am_0.9M_sample100k
18
 
19
+ This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on the mlfoundations-dev/AM-DeepSeek-R1-Distilled-1.4M-am_0.9M_sample100k dataset.
20
 
21
  ## Model description
22
 
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.950108459869848,
3
+ "total_flos": 2.1615691655827948e+18,
4
+ "train_loss": 0.9045009151810691,
5
+ "train_runtime": 13253.2478,
6
+ "train_samples_per_second": 3.337,
7
+ "train_steps_per_second": 0.006
8
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.950108459869848,
3
+ "total_flos": 2.1615691655827948e+18,
4
+ "train_loss": 0.9045009151810691,
5
+ "train_runtime": 13253.2478,
6
+ "train_samples_per_second": 3.337,
7
+ "train_steps_per_second": 0.006
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,630 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.950108459869848,
5
+ "eval_steps": 500,
6
+ "global_step": 84,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.03470715835140998,
13
+ "grad_norm": 6.300277096992606,
14
+ "learning_rate": 8.888888888888888e-06,
15
+ "loss": 1.1898,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.06941431670281996,
20
+ "grad_norm": 6.329585685276253,
21
+ "learning_rate": 1.7777777777777777e-05,
22
+ "loss": 1.1873,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.10412147505422993,
27
+ "grad_norm": 2.501481021126002,
28
+ "learning_rate": 2.6666666666666667e-05,
29
+ "loss": 1.0822,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.13882863340563992,
34
+ "grad_norm": 5.685457142160091,
35
+ "learning_rate": 3.555555555555555e-05,
36
+ "loss": 1.0872,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.1735357917570499,
41
+ "grad_norm": 9.472859854111334,
42
+ "learning_rate": 4.444444444444445e-05,
43
+ "loss": 1.1525,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.20824295010845986,
48
+ "grad_norm": 6.569946608517327,
49
+ "learning_rate": 5.333333333333333e-05,
50
+ "loss": 1.1009,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.24295010845986983,
55
+ "grad_norm": 3.450270474507439,
56
+ "learning_rate": 6.222222222222223e-05,
57
+ "loss": 1.0244,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.27765726681127983,
62
+ "grad_norm": 4.362696170141236,
63
+ "learning_rate": 7.11111111111111e-05,
64
+ "loss": 0.9832,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.3123644251626898,
69
+ "grad_norm": 2.502106333125194,
70
+ "learning_rate": 8e-05,
71
+ "loss": 0.949,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.3470715835140998,
76
+ "grad_norm": 4.093217491327189,
77
+ "learning_rate": 7.996491320395435e-05,
78
+ "loss": 0.9748,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.38177874186550975,
83
+ "grad_norm": 2.5858453011572746,
84
+ "learning_rate": 7.985971436998019e-05,
85
+ "loss": 0.9239,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 0.4164859002169197,
90
+ "grad_norm": 2.203682105160326,
91
+ "learning_rate": 7.968458805257913e-05,
92
+ "loss": 0.9064,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 0.4511930585683297,
97
+ "grad_norm": 1.5811510985617325,
98
+ "learning_rate": 7.94398414828202e-05,
99
+ "loss": 0.8827,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 0.48590021691973967,
104
+ "grad_norm": 2.803169201898971,
105
+ "learning_rate": 7.912590402935223e-05,
106
+ "loss": 0.892,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 0.5206073752711496,
111
+ "grad_norm": 1.693289513455263,
112
+ "learning_rate": 7.874332644514525e-05,
113
+ "loss": 0.887,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 0.5553145336225597,
118
+ "grad_norm": 1.086358307999921,
119
+ "learning_rate": 7.829277990128269e-05,
120
+ "loss": 0.8647,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 0.5900216919739696,
125
+ "grad_norm": 1.9102839923766066,
126
+ "learning_rate": 7.777505480949924e-05,
127
+ "loss": 0.8804,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 0.6247288503253796,
132
+ "grad_norm": 1.4012123751616394,
133
+ "learning_rate": 7.719105943553007e-05,
134
+ "loss": 0.8601,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 0.6594360086767896,
139
+ "grad_norm": 3775.810932593552,
140
+ "learning_rate": 7.654181830570404e-05,
141
+ "loss": 8.0147,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 0.6941431670281996,
146
+ "grad_norm": 597.812195875499,
147
+ "learning_rate": 7.582847040957652e-05,
148
+ "loss": 1.8169,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 0.7288503253796096,
153
+ "grad_norm": 5.978836278642803,
154
+ "learning_rate": 7.505226720175455e-05,
155
+ "loss": 1.0828,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 0.7635574837310195,
160
+ "grad_norm": 4.199965410848235,
161
+ "learning_rate": 7.421457040642028e-05,
162
+ "loss": 1.011,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 0.7982646420824295,
167
+ "grad_norm": 1.7290854744382491,
168
+ "learning_rate": 7.331684962840398e-05,
169
+ "loss": 0.9144,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 0.8329718004338394,
174
+ "grad_norm": 1.910958548071721,
175
+ "learning_rate": 7.236067977499791e-05,
176
+ "loss": 0.8651,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 0.8676789587852495,
181
+ "grad_norm": 1.764578753926621,
182
+ "learning_rate": 7.13477382930336e-05,
183
+ "loss": 0.8855,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 0.9023861171366594,
188
+ "grad_norm": 1.8041505018631134,
189
+ "learning_rate": 7.027980222607026e-05,
190
+ "loss": 0.8672,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 0.9370932754880694,
195
+ "grad_norm": 1.3191472044863197,
196
+ "learning_rate": 6.915874509685646e-05,
197
+ "loss": 0.8423,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 0.9718004338394793,
202
+ "grad_norm": 0.9189517507692654,
203
+ "learning_rate": 6.798653362053462e-05,
204
+ "loss": 0.8273,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 1.0238611713665944,
209
+ "grad_norm": 2.1289244779300835,
210
+ "learning_rate": 6.676522425435433e-05,
211
+ "loss": 1.3774,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 1.0585683297180044,
216
+ "grad_norm": 25.961017896043277,
217
+ "learning_rate": 6.54969595899476e-05,
218
+ "loss": 0.7943,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 1.0932754880694142,
223
+ "grad_norm": 1.529851042182263,
224
+ "learning_rate": 6.418396459449501e-05,
225
+ "loss": 0.8102,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 1.1279826464208242,
230
+ "grad_norm": 8.191973267629333,
231
+ "learning_rate": 6.282854270737728e-05,
232
+ "loss": 0.8152,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 1.1626898047722343,
237
+ "grad_norm": 1.209748771757014,
238
+ "learning_rate": 6.143307179915987e-05,
239
+ "loss": 0.7962,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 1.1973969631236443,
244
+ "grad_norm": 0.7840741589903159,
245
+ "learning_rate": 6.000000000000001e-05,
246
+ "loss": 0.7816,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 1.2321041214750543,
251
+ "grad_norm": 0.9740463452743966,
252
+ "learning_rate": 5.853184140479448e-05,
253
+ "loss": 0.7605,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 1.2668112798264641,
258
+ "grad_norm": 0.7045833732067921,
259
+ "learning_rate": 5.703117166260291e-05,
260
+ "loss": 0.7669,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 1.3015184381778742,
265
+ "grad_norm": 0.8969503221689408,
266
+ "learning_rate": 5.5500623458084123e-05,
267
+ "loss": 0.7662,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 1.3362255965292842,
272
+ "grad_norm": 0.6025365616410666,
273
+ "learning_rate": 5.394288189287261e-05,
274
+ "loss": 0.7564,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 1.3709327548806942,
279
+ "grad_norm": 0.7688283335699936,
280
+ "learning_rate": 5.23606797749979e-05,
281
+ "loss": 0.7631,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 1.405639913232104,
286
+ "grad_norm": 0.48261395895365083,
287
+ "learning_rate": 5.075679282461063e-05,
288
+ "loss": 0.7557,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 1.440347071583514,
293
+ "grad_norm": 0.5414576203146455,
294
+ "learning_rate": 4.9134034804426235e-05,
295
+ "loss": 0.741,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 1.475054229934924,
300
+ "grad_norm": 0.4233507247912478,
301
+ "learning_rate": 4.7495252583429e-05,
302
+ "loss": 0.7463,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 1.509761388286334,
307
+ "grad_norm": 0.37870693147142465,
308
+ "learning_rate": 4.5843321142496474e-05,
309
+ "loss": 0.7457,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 1.5444685466377441,
314
+ "grad_norm": 0.4059373324038044,
315
+ "learning_rate": 4.418113853070614e-05,
316
+ "loss": 0.7411,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 1.579175704989154,
321
+ "grad_norm": 0.33414673476249707,
322
+ "learning_rate": 4.251162078117254e-05,
323
+ "loss": 0.7298,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 1.613882863340564,
328
+ "grad_norm": 0.3398435439622563,
329
+ "learning_rate": 4.0837696795334284e-05,
330
+ "loss": 0.7371,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 1.648590021691974,
335
+ "grad_norm": 0.37740119571438335,
336
+ "learning_rate": 3.916230320466573e-05,
337
+ "loss": 0.7303,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 1.6832971800433838,
342
+ "grad_norm": 0.3247066964457289,
343
+ "learning_rate": 3.7488379218827466e-05,
344
+ "loss": 0.7386,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 1.718004338394794,
349
+ "grad_norm": 0.3379048377668955,
350
+ "learning_rate": 3.581886146929387e-05,
351
+ "loss": 0.7259,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 1.7527114967462039,
356
+ "grad_norm": 0.3181669649842156,
357
+ "learning_rate": 3.415667885750354e-05,
358
+ "loss": 0.7209,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 1.7874186550976139,
363
+ "grad_norm": 0.32411152865418363,
364
+ "learning_rate": 3.250474741657101e-05,
365
+ "loss": 0.7251,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 1.822125813449024,
370
+ "grad_norm": 0.23388485665108716,
371
+ "learning_rate": 3.086596519557377e-05,
372
+ "loss": 0.723,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 1.8568329718004337,
377
+ "grad_norm": 0.3048906277879464,
378
+ "learning_rate": 2.9243207175389377e-05,
379
+ "loss": 0.7245,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 1.891540130151844,
384
+ "grad_norm": 0.21182971362551556,
385
+ "learning_rate": 2.7639320225002108e-05,
386
+ "loss": 0.7262,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 1.9262472885032538,
391
+ "grad_norm": 0.25916446139863025,
392
+ "learning_rate": 2.6057118107127402e-05,
393
+ "loss": 0.7228,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 1.9609544468546638,
398
+ "grad_norm": 0.17974296947054186,
399
+ "learning_rate": 2.4499376541915883e-05,
400
+ "loss": 0.7208,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 2.013015184381779,
405
+ "grad_norm": 0.36891336506559885,
406
+ "learning_rate": 2.2968828337397095e-05,
407
+ "loss": 1.1849,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 2.0477223427331888,
412
+ "grad_norm": 0.220956290587299,
413
+ "learning_rate": 2.1468158595205543e-05,
414
+ "loss": 0.6845,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 2.0824295010845986,
419
+ "grad_norm": 0.19743211488551957,
420
+ "learning_rate": 2.0000000000000012e-05,
421
+ "loss": 0.681,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 2.117136659436009,
426
+ "grad_norm": 0.22014276891240445,
427
+ "learning_rate": 1.8566928200840128e-05,
428
+ "loss": 0.6803,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 2.1518438177874186,
433
+ "grad_norm": 0.24004715720654304,
434
+ "learning_rate": 1.7171457292622736e-05,
435
+ "loss": 0.6756,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 2.1865509761388284,
440
+ "grad_norm": 0.16300098008860384,
441
+ "learning_rate": 1.5816035405505007e-05,
442
+ "loss": 0.6713,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 2.2212581344902387,
447
+ "grad_norm": 0.2029252686694922,
448
+ "learning_rate": 1.4503040410052412e-05,
449
+ "loss": 0.6778,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 2.2559652928416485,
454
+ "grad_norm": 0.16175699949865352,
455
+ "learning_rate": 1.3234775745645684e-05,
456
+ "loss": 0.6733,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 2.2906724511930587,
461
+ "grad_norm": 0.15518204077737663,
462
+ "learning_rate": 1.201346637946538e-05,
463
+ "loss": 0.673,
464
+ "step": 65
465
+ },
466
+ {
467
+ "epoch": 2.3253796095444685,
468
+ "grad_norm": 0.15484153932844605,
469
+ "learning_rate": 1.0841254903143547e-05,
470
+ "loss": 0.6758,
471
+ "step": 66
472
+ },
473
+ {
474
+ "epoch": 2.3600867678958783,
475
+ "grad_norm": 0.1425511720931603,
476
+ "learning_rate": 9.720197773929749e-06,
477
+ "loss": 0.6708,
478
+ "step": 67
479
+ },
480
+ {
481
+ "epoch": 2.3947939262472886,
482
+ "grad_norm": 0.1315080719342115,
483
+ "learning_rate": 8.652261706966412e-06,
484
+ "loss": 0.6694,
485
+ "step": 68
486
+ },
487
+ {
488
+ "epoch": 2.4295010845986984,
489
+ "grad_norm": 0.13003673674800192,
490
+ "learning_rate": 7.639320225002106e-06,
491
+ "loss": 0.6749,
492
+ "step": 69
493
+ },
494
+ {
495
+ "epoch": 2.4642082429501087,
496
+ "grad_norm": 0.12937383195418148,
497
+ "learning_rate": 6.6831503715960235e-06,
498
+ "loss": 0.6752,
499
+ "step": 70
500
+ },
501
+ {
502
+ "epoch": 2.4989154013015185,
503
+ "grad_norm": 0.12141271020145257,
504
+ "learning_rate": 5.785429593579736e-06,
505
+ "loss": 0.6781,
506
+ "step": 71
507
+ },
508
+ {
509
+ "epoch": 2.5336225596529283,
510
+ "grad_norm": 0.11519870379159697,
511
+ "learning_rate": 4.947732798245466e-06,
512
+ "loss": 0.6716,
513
+ "step": 72
514
+ },
515
+ {
516
+ "epoch": 2.5683297180043385,
517
+ "grad_norm": 0.11773898061730143,
518
+ "learning_rate": 4.171529590423484e-06,
519
+ "loss": 0.6779,
520
+ "step": 73
521
+ },
522
+ {
523
+ "epoch": 2.6030368763557483,
524
+ "grad_norm": 0.10608208259960443,
525
+ "learning_rate": 3.458181694295961e-06,
526
+ "loss": 0.6805,
527
+ "step": 74
528
+ },
529
+ {
530
+ "epoch": 2.6377440347071586,
531
+ "grad_norm": 0.10236639050094021,
532
+ "learning_rate": 2.8089405644699463e-06,
533
+ "loss": 0.6719,
534
+ "step": 75
535
+ },
536
+ {
537
+ "epoch": 2.6724511930585684,
538
+ "grad_norm": 0.10603330100678318,
539
+ "learning_rate": 2.2249451905007603e-06,
540
+ "loss": 0.675,
541
+ "step": 76
542
+ },
543
+ {
544
+ "epoch": 2.707158351409978,
545
+ "grad_norm": 0.09832873419536761,
546
+ "learning_rate": 1.7072200987173105e-06,
547
+ "loss": 0.6767,
548
+ "step": 77
549
+ },
550
+ {
551
+ "epoch": 2.7418655097613884,
552
+ "grad_norm": 0.0983832265038647,
553
+ "learning_rate": 1.256673554854757e-06,
554
+ "loss": 0.6633,
555
+ "step": 78
556
+ },
557
+ {
558
+ "epoch": 2.7765726681127982,
559
+ "grad_norm": 0.10299057334518875,
560
+ "learning_rate": 8.740959706477725e-07,
561
+ "loss": 0.6672,
562
+ "step": 79
563
+ },
564
+ {
565
+ "epoch": 2.811279826464208,
566
+ "grad_norm": 0.09743605608109047,
567
+ "learning_rate": 5.601585171798052e-07,
568
+ "loss": 0.6708,
569
+ "step": 80
570
+ },
571
+ {
572
+ "epoch": 2.8459869848156183,
573
+ "grad_norm": 0.10075730943046303,
574
+ "learning_rate": 3.154119474208894e-07,
575
+ "loss": 0.6748,
576
+ "step": 81
577
+ },
578
+ {
579
+ "epoch": 2.880694143167028,
580
+ "grad_norm": 0.09556334296668786,
581
+ "learning_rate": 1.402856300198252e-07,
582
+ "loss": 0.6686,
583
+ "step": 82
584
+ },
585
+ {
586
+ "epoch": 2.915401301518438,
587
+ "grad_norm": 0.09223259109939756,
588
+ "learning_rate": 3.5086796045664674e-08,
589
+ "loss": 0.662,
590
+ "step": 83
591
+ },
592
+ {
593
+ "epoch": 2.950108459869848,
594
+ "grad_norm": 0.09460606099922247,
595
+ "learning_rate": 0.0,
596
+ "loss": 0.673,
597
+ "step": 84
598
+ },
599
+ {
600
+ "epoch": 2.950108459869848,
601
+ "step": 84,
602
+ "total_flos": 2.1615691655827948e+18,
603
+ "train_loss": 0.9045009151810691,
604
+ "train_runtime": 13253.2478,
605
+ "train_samples_per_second": 3.337,
606
+ "train_steps_per_second": 0.006
607
+ }
608
+ ],
609
+ "logging_steps": 1.0,
610
+ "max_steps": 84,
611
+ "num_input_tokens_seen": 0,
612
+ "num_train_epochs": 3,
613
+ "save_steps": 500,
614
+ "stateful_callbacks": {
615
+ "TrainerControl": {
616
+ "args": {
617
+ "should_epoch_stop": false,
618
+ "should_evaluate": false,
619
+ "should_log": false,
620
+ "should_save": true,
621
+ "should_training_stop": true
622
+ },
623
+ "attributes": {}
624
+ }
625
+ },
626
+ "total_flos": 2.1615691655827948e+18,
627
+ "train_batch_size": 1,
628
+ "trial_name": null,
629
+ "trial_params": null
630
+ }
training_loss.png ADDED