furmaniak commited on
Commit
4a2d6c5
·
verified ·
1 Parent(s): 208354a

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 +861 -0
  5. training_loss.png +0 -0
README.md CHANGED
@@ -4,6 +4,7 @@ license: mit
4
  base_model: microsoft/phi-4
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: phi-4-openalex
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # phi-4-openalex
17
 
18
- This model is a fine-tuned version of [microsoft/phi-4](https://huggingface.co/microsoft/phi-4) on an unknown dataset.
19
 
20
  ## Model description
21
 
 
4
  base_model: microsoft/phi-4
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: phi-4-openalex
 
16
 
17
  # phi-4-openalex
18
 
19
+ This model is a fine-tuned version of [microsoft/phi-4](https://huggingface.co/microsoft/phi-4) on the openalex dataset.
20
 
21
  ## Model description
22
 
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 609898777804800.0,
4
+ "train_loss": 1.8114001343392918,
5
+ "train_runtime": 22537.3565,
6
+ "train_samples_per_second": 1.328,
7
+ "train_steps_per_second": 0.005
8
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 609898777804800.0,
4
+ "train_loss": 1.8114001343392918,
5
+ "train_runtime": 22537.3565,
6
+ "train_samples_per_second": 1.328,
7
+ "train_steps_per_second": 0.005
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,861 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 117,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.008547008547008548,
13
+ "grad_norm": 0.13582941889762878,
14
+ "learning_rate": 1.6666666666666667e-06,
15
+ "loss": 1.8884,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.017094017094017096,
20
+ "grad_norm": 0.13621631264686584,
21
+ "learning_rate": 3.3333333333333333e-06,
22
+ "loss": 1.8929,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.02564102564102564,
27
+ "grad_norm": 0.11699245870113373,
28
+ "learning_rate": 5e-06,
29
+ "loss": 1.9025,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.03418803418803419,
34
+ "grad_norm": 0.05705893784761429,
35
+ "learning_rate": 6.666666666666667e-06,
36
+ "loss": 1.8795,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.042735042735042736,
41
+ "grad_norm": 0.056257281452417374,
42
+ "learning_rate": 8.333333333333334e-06,
43
+ "loss": 1.8705,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.05128205128205128,
48
+ "grad_norm": 0.04789767786860466,
49
+ "learning_rate": 1e-05,
50
+ "loss": 1.8572,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.05982905982905983,
55
+ "grad_norm": 0.08197996765375137,
56
+ "learning_rate": 9.99799753559161e-06,
57
+ "loss": 1.8696,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.06837606837606838,
62
+ "grad_norm": 0.05935126915574074,
63
+ "learning_rate": 9.991991746311916e-06,
64
+ "loss": 1.8535,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.07692307692307693,
69
+ "grad_norm": 0.08124881237745285,
70
+ "learning_rate": 9.981987442712634e-06,
71
+ "loss": 1.8623,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.08547008547008547,
76
+ "grad_norm": 0.038786403834819794,
77
+ "learning_rate": 9.967992638098517e-06,
78
+ "loss": 1.8575,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.09401709401709402,
83
+ "grad_norm": 0.061246760189533234,
84
+ "learning_rate": 9.950018542108818e-06,
85
+ "loss": 1.8533,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 0.10256410256410256,
90
+ "grad_norm": 0.052762433886528015,
91
+ "learning_rate": 9.928079551738542e-06,
92
+ "loss": 1.8447,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 0.1111111111111111,
97
+ "grad_norm": 0.045264095067977905,
98
+ "learning_rate": 9.902193239806634e-06,
99
+ "loss": 1.8458,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 0.11965811965811966,
104
+ "grad_norm": 0.049116991460323334,
105
+ "learning_rate": 9.872380340880416e-06,
106
+ "loss": 1.8447,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 0.1282051282051282,
111
+ "grad_norm": 0.04788421094417572,
112
+ "learning_rate": 9.838664734667496e-06,
113
+ "loss": 1.8474,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 0.13675213675213677,
118
+ "grad_norm": 0.03990117087960243,
119
+ "learning_rate": 9.801073426888447e-06,
120
+ "loss": 1.8424,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 0.1452991452991453,
125
+ "grad_norm": 0.0356304794549942,
126
+ "learning_rate": 9.759636527645633e-06,
127
+ "loss": 1.8274,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 0.15384615384615385,
132
+ "grad_norm": 0.04112239554524422,
133
+ "learning_rate": 9.714387227305422e-06,
134
+ "loss": 1.8262,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 0.1623931623931624,
139
+ "grad_norm": 0.04289136081933975,
140
+ "learning_rate": 9.665361769913187e-06,
141
+ "loss": 1.8331,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 0.17094017094017094,
146
+ "grad_norm": 0.03238912671804428,
147
+ "learning_rate": 9.612599424162344e-06,
148
+ "loss": 1.8278,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 0.1794871794871795,
153
+ "grad_norm": 0.03170251473784447,
154
+ "learning_rate": 9.55614245194068e-06,
155
+ "loss": 1.8238,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 0.18803418803418803,
160
+ "grad_norm": 0.03161657974123955,
161
+ "learning_rate": 9.496036074479184e-06,
162
+ "loss": 1.8282,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 0.19658119658119658,
167
+ "grad_norm": 0.028967849910259247,
168
+ "learning_rate": 9.432328436130493e-06,
169
+ "loss": 1.825,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 0.20512820512820512,
174
+ "grad_norm": 0.025380073115229607,
175
+ "learning_rate": 9.365070565805941e-06,
176
+ "loss": 1.8165,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 0.21367521367521367,
181
+ "grad_norm": 0.027646927163004875,
182
+ "learning_rate": 9.294316336102132e-06,
183
+ "loss": 1.8194,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 0.2222222222222222,
188
+ "grad_norm": 0.02571098506450653,
189
+ "learning_rate": 9.220122420149753e-06,
190
+ "loss": 1.822,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 0.23076923076923078,
195
+ "grad_norm": 0.02322917804121971,
196
+ "learning_rate": 9.142548246219212e-06,
197
+ "loss": 1.8218,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 0.23931623931623933,
202
+ "grad_norm": 0.022094452753663063,
203
+ "learning_rate": 9.06165595011943e-06,
204
+ "loss": 1.8245,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 0.24786324786324787,
209
+ "grad_norm": 0.025401204824447632,
210
+ "learning_rate": 8.97751032542795e-06,
211
+ "loss": 1.8087,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 0.2564102564102564,
216
+ "grad_norm": 0.02527693286538124,
217
+ "learning_rate": 8.890178771592198e-06,
218
+ "loss": 1.8097,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 0.26495726495726496,
223
+ "grad_norm": 0.02179970219731331,
224
+ "learning_rate": 8.799731239943488e-06,
225
+ "loss": 1.8207,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 0.27350427350427353,
230
+ "grad_norm": 0.01948496885597706,
231
+ "learning_rate": 8.706240177667003e-06,
232
+ "loss": 1.8124,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 0.28205128205128205,
237
+ "grad_norm": 0.01962677575647831,
238
+ "learning_rate": 8.609780469772623e-06,
239
+ "loss": 1.8184,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 0.2905982905982906,
244
+ "grad_norm": 0.019485613331198692,
245
+ "learning_rate": 8.510429379113114e-06,
246
+ "loss": 1.819,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 0.29914529914529914,
251
+ "grad_norm": 0.01905289851129055,
252
+ "learning_rate": 8.408266484497664e-06,
253
+ "loss": 1.8153,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 0.3076923076923077,
258
+ "grad_norm": 0.019722452387213707,
259
+ "learning_rate": 8.303373616950408e-06,
260
+ "loss": 1.8141,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 0.3162393162393162,
265
+ "grad_norm": 0.02127656526863575,
266
+ "learning_rate": 8.195834794164925e-06,
267
+ "loss": 1.8072,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 0.3247863247863248,
272
+ "grad_norm": 0.020507801324129105,
273
+ "learning_rate": 8.085736153207277e-06,
274
+ "loss": 1.8171,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 0.3333333333333333,
279
+ "grad_norm": 0.016468843445181847,
280
+ "learning_rate": 7.973165881521435e-06,
281
+ "loss": 1.8192,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 0.3418803418803419,
286
+ "grad_norm": 0.01777919940650463,
287
+ "learning_rate": 7.858214146292394e-06,
288
+ "loss": 1.8069,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 0.3504273504273504,
293
+ "grad_norm": 0.01822058856487274,
294
+ "learning_rate": 7.74097302222355e-06,
295
+ "loss": 1.8093,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 0.358974358974359,
300
+ "grad_norm": 0.0196178387850523,
301
+ "learning_rate": 7.621536417786159e-06,
302
+ "loss": 1.8014,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 0.36752136752136755,
307
+ "grad_norm": 0.017990173771977425,
308
+ "learning_rate": 7.500000000000001e-06,
309
+ "loss": 1.8108,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 0.37606837606837606,
314
+ "grad_norm": 0.01827268861234188,
315
+ "learning_rate": 7.37646111780545e-06,
316
+ "loss": 1.8069,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 0.38461538461538464,
321
+ "grad_norm": 0.016010349616408348,
322
+ "learning_rate": 7.251018724088367e-06,
323
+ "loss": 1.8057,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 0.39316239316239315,
328
+ "grad_norm": 0.01656198874115944,
329
+ "learning_rate": 7.12377329642024e-06,
330
+ "loss": 1.8065,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 0.4017094017094017,
335
+ "grad_norm": 0.01725705899298191,
336
+ "learning_rate": 6.994826756577082e-06,
337
+ "loss": 1.7997,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 0.41025641025641024,
342
+ "grad_norm": 0.01705297827720642,
343
+ "learning_rate": 6.864282388901544e-06,
344
+ "loss": 1.8073,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 0.4188034188034188,
349
+ "grad_norm": 0.015689667314291,
350
+ "learning_rate": 6.732244757573619e-06,
351
+ "loss": 1.8056,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 0.42735042735042733,
356
+ "grad_norm": 0.01496344804763794,
357
+ "learning_rate": 6.598819622856227e-06,
358
+ "loss": 1.807,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 0.4358974358974359,
363
+ "grad_norm": 0.015072242356836796,
364
+ "learning_rate": 6.464113856382752e-06,
365
+ "loss": 1.8098,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 0.4444444444444444,
370
+ "grad_norm": 0.015582077205181122,
371
+ "learning_rate": 6.328235355554382e-06,
372
+ "loss": 1.8016,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 0.452991452991453,
377
+ "grad_norm": 0.015510096214711666,
378
+ "learning_rate": 6.191292957115825e-06,
379
+ "loss": 1.8111,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 0.46153846153846156,
384
+ "grad_norm": 0.01476968452334404,
385
+ "learning_rate": 6.053396349978632e-06,
386
+ "loss": 1.8038,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 0.4700854700854701,
391
+ "grad_norm": 0.014734070748090744,
392
+ "learning_rate": 5.914655987361934e-06,
393
+ "loss": 1.7871,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 0.47863247863247865,
398
+ "grad_norm": 0.015119190327823162,
399
+ "learning_rate": 5.77518299832099e-06,
400
+ "loss": 1.8018,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 0.48717948717948717,
405
+ "grad_norm": 0.01506562065333128,
406
+ "learning_rate": 5.635089098734394e-06,
407
+ "loss": 1.8044,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 0.49572649572649574,
412
+ "grad_norm": 0.014175213873386383,
413
+ "learning_rate": 5.49448650182125e-06,
414
+ "loss": 1.8005,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 0.5042735042735043,
419
+ "grad_norm": 0.014436254277825356,
420
+ "learning_rate": 5.353487828259973e-06,
421
+ "loss": 1.8026,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 0.5128205128205128,
426
+ "grad_norm": 0.014683015644550323,
427
+ "learning_rate": 5.212206015980742e-06,
428
+ "loss": 1.7954,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 0.5213675213675214,
433
+ "grad_norm": 0.014612887986004353,
434
+ "learning_rate": 5.070754229703811e-06,
435
+ "loss": 1.7961,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 0.5299145299145299,
440
+ "grad_norm": 0.014702665619552135,
441
+ "learning_rate": 4.929245770296191e-06,
442
+ "loss": 1.8036,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 0.5384615384615384,
447
+ "grad_norm": 0.014264926314353943,
448
+ "learning_rate": 4.78779398401926e-06,
449
+ "loss": 1.8029,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 0.5470085470085471,
454
+ "grad_norm": 0.014685841277241707,
455
+ "learning_rate": 4.646512171740028e-06,
456
+ "loss": 1.8026,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 0.5555555555555556,
461
+ "grad_norm": 0.01473164837807417,
462
+ "learning_rate": 4.505513498178752e-06,
463
+ "loss": 1.8079,
464
+ "step": 65
465
+ },
466
+ {
467
+ "epoch": 0.5641025641025641,
468
+ "grad_norm": 0.014097092673182487,
469
+ "learning_rate": 4.364910901265607e-06,
470
+ "loss": 1.8008,
471
+ "step": 66
472
+ },
473
+ {
474
+ "epoch": 0.5726495726495726,
475
+ "grad_norm": 0.014091964811086655,
476
+ "learning_rate": 4.224817001679011e-06,
477
+ "loss": 1.7983,
478
+ "step": 67
479
+ },
480
+ {
481
+ "epoch": 0.5811965811965812,
482
+ "grad_norm": 0.01420042384415865,
483
+ "learning_rate": 4.085344012638067e-06,
484
+ "loss": 1.799,
485
+ "step": 68
486
+ },
487
+ {
488
+ "epoch": 0.5897435897435898,
489
+ "grad_norm": 0.01425846852362156,
490
+ "learning_rate": 3.94660365002137e-06,
491
+ "loss": 1.7981,
492
+ "step": 69
493
+ },
494
+ {
495
+ "epoch": 0.5982905982905983,
496
+ "grad_norm": 0.1001775860786438,
497
+ "learning_rate": 3.808707042884176e-06,
498
+ "loss": 1.7983,
499
+ "step": 70
500
+ },
501
+ {
502
+ "epoch": 0.6068376068376068,
503
+ "grad_norm": 0.014455100521445274,
504
+ "learning_rate": 3.6717646444456196e-06,
505
+ "loss": 1.7924,
506
+ "step": 71
507
+ },
508
+ {
509
+ "epoch": 0.6153846153846154,
510
+ "grad_norm": 0.014086514711380005,
511
+ "learning_rate": 3.5358861436172487e-06,
512
+ "loss": 1.799,
513
+ "step": 72
514
+ },
515
+ {
516
+ "epoch": 0.6239316239316239,
517
+ "grad_norm": 0.015152843669056892,
518
+ "learning_rate": 3.401180377143774e-06,
519
+ "loss": 1.8005,
520
+ "step": 73
521
+ },
522
+ {
523
+ "epoch": 0.6324786324786325,
524
+ "grad_norm": 0.014293001964688301,
525
+ "learning_rate": 3.2677552424263836e-06,
526
+ "loss": 1.7956,
527
+ "step": 74
528
+ },
529
+ {
530
+ "epoch": 0.6410256410256411,
531
+ "grad_norm": 0.014682993292808533,
532
+ "learning_rate": 3.1357176110984578e-06,
533
+ "loss": 1.798,
534
+ "step": 75
535
+ },
536
+ {
537
+ "epoch": 0.6495726495726496,
538
+ "grad_norm": 0.014120382256805897,
539
+ "learning_rate": 3.0051732434229185e-06,
540
+ "loss": 1.7934,
541
+ "step": 76
542
+ },
543
+ {
544
+ "epoch": 0.6581196581196581,
545
+ "grad_norm": 0.014230918139219284,
546
+ "learning_rate": 2.8762267035797607e-06,
547
+ "loss": 1.8012,
548
+ "step": 77
549
+ },
550
+ {
551
+ "epoch": 0.6666666666666666,
552
+ "grad_norm": 0.01403116900473833,
553
+ "learning_rate": 2.748981275911633e-06,
554
+ "loss": 1.7915,
555
+ "step": 78
556
+ },
557
+ {
558
+ "epoch": 0.6752136752136753,
559
+ "grad_norm": 0.013814397156238556,
560
+ "learning_rate": 2.6235388821945497e-06,
561
+ "loss": 1.7867,
562
+ "step": 79
563
+ },
564
+ {
565
+ "epoch": 0.6837606837606838,
566
+ "grad_norm": 0.013797773979604244,
567
+ "learning_rate": 2.5000000000000015e-06,
568
+ "loss": 1.7952,
569
+ "step": 80
570
+ },
571
+ {
572
+ "epoch": 0.6923076923076923,
573
+ "grad_norm": 0.013870066963136196,
574
+ "learning_rate": 2.3784635822138424e-06,
575
+ "loss": 1.8026,
576
+ "step": 81
577
+ },
578
+ {
579
+ "epoch": 0.7008547008547008,
580
+ "grad_norm": 0.014055611565709114,
581
+ "learning_rate": 2.2590269777764516e-06,
582
+ "loss": 1.7956,
583
+ "step": 82
584
+ },
585
+ {
586
+ "epoch": 0.7094017094017094,
587
+ "grad_norm": 0.013872554525732994,
588
+ "learning_rate": 2.141785853707607e-06,
589
+ "loss": 1.7947,
590
+ "step": 83
591
+ },
592
+ {
593
+ "epoch": 0.717948717948718,
594
+ "grad_norm": 0.013987288810312748,
595
+ "learning_rate": 2.0268341184785674e-06,
596
+ "loss": 1.7963,
597
+ "step": 84
598
+ },
599
+ {
600
+ "epoch": 0.7264957264957265,
601
+ "grad_norm": 0.0137988505885005,
602
+ "learning_rate": 1.9142638467927254e-06,
603
+ "loss": 1.7974,
604
+ "step": 85
605
+ },
606
+ {
607
+ "epoch": 0.7350427350427351,
608
+ "grad_norm": 0.013994401320815086,
609
+ "learning_rate": 1.8041652058350768e-06,
610
+ "loss": 1.794,
611
+ "step": 86
612
+ },
613
+ {
614
+ "epoch": 0.7435897435897436,
615
+ "grad_norm": 0.013571497052907944,
616
+ "learning_rate": 1.6966263830495939e-06,
617
+ "loss": 1.8017,
618
+ "step": 87
619
+ },
620
+ {
621
+ "epoch": 0.7521367521367521,
622
+ "grad_norm": 0.018896423280239105,
623
+ "learning_rate": 1.5917335155023368e-06,
624
+ "loss": 1.7874,
625
+ "step": 88
626
+ },
627
+ {
628
+ "epoch": 0.7606837606837606,
629
+ "grad_norm": 0.013724449090659618,
630
+ "learning_rate": 1.4895706208868876e-06,
631
+ "loss": 1.8012,
632
+ "step": 89
633
+ },
634
+ {
635
+ "epoch": 0.7692307692307693,
636
+ "grad_norm": 0.014079892076551914,
637
+ "learning_rate": 1.390219530227378e-06,
638
+ "loss": 1.8029,
639
+ "step": 90
640
+ },
641
+ {
642
+ "epoch": 0.7777777777777778,
643
+ "grad_norm": 0.01355001050978899,
644
+ "learning_rate": 1.2937598223330006e-06,
645
+ "loss": 1.7955,
646
+ "step": 91
647
+ },
648
+ {
649
+ "epoch": 0.7863247863247863,
650
+ "grad_norm": 0.013746261596679688,
651
+ "learning_rate": 1.2002687600565138e-06,
652
+ "loss": 1.7919,
653
+ "step": 92
654
+ },
655
+ {
656
+ "epoch": 0.7948717948717948,
657
+ "grad_norm": 0.013557142578065395,
658
+ "learning_rate": 1.1098212284078037e-06,
659
+ "loss": 1.794,
660
+ "step": 93
661
+ },
662
+ {
663
+ "epoch": 0.8034188034188035,
664
+ "grad_norm": 0.014017937704920769,
665
+ "learning_rate": 1.0224896745720513e-06,
666
+ "loss": 1.7978,
667
+ "step": 94
668
+ },
669
+ {
670
+ "epoch": 0.811965811965812,
671
+ "grad_norm": 0.013620711863040924,
672
+ "learning_rate": 9.383440498805712e-07,
673
+ "loss": 1.8013,
674
+ "step": 95
675
+ },
676
+ {
677
+ "epoch": 0.8205128205128205,
678
+ "grad_norm": 0.013980009593069553,
679
+ "learning_rate": 8.574517537807897e-07,
680
+ "loss": 1.7928,
681
+ "step": 96
682
+ },
683
+ {
684
+ "epoch": 0.8290598290598291,
685
+ "grad_norm": 0.013654535636305809,
686
+ "learning_rate": 7.798775798502484e-07,
687
+ "loss": 1.7875,
688
+ "step": 97
689
+ },
690
+ {
691
+ "epoch": 0.8376068376068376,
692
+ "grad_norm": 0.013558434322476387,
693
+ "learning_rate": 7.056836638978698e-07,
694
+ "loss": 1.7868,
695
+ "step": 98
696
+ },
697
+ {
698
+ "epoch": 0.8461538461538461,
699
+ "grad_norm": 0.01349763572216034,
700
+ "learning_rate": 6.349294341940593e-07,
701
+ "loss": 1.7947,
702
+ "step": 99
703
+ },
704
+ {
705
+ "epoch": 0.8547008547008547,
706
+ "grad_norm": 0.013618433848023415,
707
+ "learning_rate": 5.676715638695063e-07,
708
+ "loss": 1.7933,
709
+ "step": 100
710
+ },
711
+ {
712
+ "epoch": 0.8632478632478633,
713
+ "grad_norm": 0.013701952062547207,
714
+ "learning_rate": 5.039639255208156e-07,
715
+ "loss": 1.7983,
716
+ "step": 101
717
+ },
718
+ {
719
+ "epoch": 0.8717948717948718,
720
+ "grad_norm": 0.013862798921763897,
721
+ "learning_rate": 4.43857548059321e-07,
722
+ "loss": 1.7923,
723
+ "step": 102
724
+ },
725
+ {
726
+ "epoch": 0.8803418803418803,
727
+ "grad_norm": 0.013503090478479862,
728
+ "learning_rate": 3.87400575837657e-07,
729
+ "loss": 1.8015,
730
+ "step": 103
731
+ },
732
+ {
733
+ "epoch": 0.8888888888888888,
734
+ "grad_norm": 0.01364339329302311,
735
+ "learning_rate": 3.346382300868134e-07,
736
+ "loss": 1.7987,
737
+ "step": 104
738
+ },
739
+ {
740
+ "epoch": 0.8974358974358975,
741
+ "grad_norm": 0.013393942266702652,
742
+ "learning_rate": 2.85612772694579e-07,
743
+ "loss": 1.801,
744
+ "step": 105
745
+ },
746
+ {
747
+ "epoch": 0.905982905982906,
748
+ "grad_norm": 0.01394713670015335,
749
+ "learning_rate": 2.403634723543674e-07,
750
+ "loss": 1.797,
751
+ "step": 106
752
+ },
753
+ {
754
+ "epoch": 0.9145299145299145,
755
+ "grad_norm": 0.013658151961863041,
756
+ "learning_rate": 1.989265731115525e-07,
757
+ "loss": 1.7963,
758
+ "step": 107
759
+ },
760
+ {
761
+ "epoch": 0.9230769230769231,
762
+ "grad_norm": 0.013742955401539803,
763
+ "learning_rate": 1.6133526533250566e-07,
764
+ "loss": 1.7919,
765
+ "step": 108
766
+ },
767
+ {
768
+ "epoch": 0.9316239316239316,
769
+ "grad_norm": 0.013528905808925629,
770
+ "learning_rate": 1.2761965911958385e-07,
771
+ "loss": 1.7973,
772
+ "step": 109
773
+ },
774
+ {
775
+ "epoch": 0.9401709401709402,
776
+ "grad_norm": 0.01358366571366787,
777
+ "learning_rate": 9.780676019336632e-08,
778
+ "loss": 1.7944,
779
+ "step": 110
780
+ },
781
+ {
782
+ "epoch": 0.9487179487179487,
783
+ "grad_norm": 0.017376506701111794,
784
+ "learning_rate": 7.192044826145772e-08,
785
+ "loss": 1.7929,
786
+ "step": 111
787
+ },
788
+ {
789
+ "epoch": 0.9572649572649573,
790
+ "grad_norm": 0.013746129348874092,
791
+ "learning_rate": 4.998145789118114e-08,
792
+ "loss": 1.7953,
793
+ "step": 112
794
+ },
795
+ {
796
+ "epoch": 0.9658119658119658,
797
+ "grad_norm": 0.013639306649565697,
798
+ "learning_rate": 3.2007361901485455e-08,
799
+ "loss": 1.794,
800
+ "step": 113
801
+ },
802
+ {
803
+ "epoch": 0.9743589743589743,
804
+ "grad_norm": 0.013497625477612019,
805
+ "learning_rate": 1.8012557287367394e-08,
806
+ "loss": 1.7915,
807
+ "step": 114
808
+ },
809
+ {
810
+ "epoch": 0.9829059829059829,
811
+ "grad_norm": 0.013749259524047375,
812
+ "learning_rate": 8.008253688084888e-09,
813
+ "loss": 1.7899,
814
+ "step": 115
815
+ },
816
+ {
817
+ "epoch": 0.9914529914529915,
818
+ "grad_norm": 0.013962676748633385,
819
+ "learning_rate": 2.002464408392135e-09,
820
+ "loss": 1.7893,
821
+ "step": 116
822
+ },
823
+ {
824
+ "epoch": 1.0,
825
+ "grad_norm": 0.013671280816197395,
826
+ "learning_rate": 0.0,
827
+ "loss": 1.7864,
828
+ "step": 117
829
+ },
830
+ {
831
+ "epoch": 1.0,
832
+ "step": 117,
833
+ "total_flos": 609898777804800.0,
834
+ "train_loss": 1.8114001343392918,
835
+ "train_runtime": 22537.3565,
836
+ "train_samples_per_second": 1.328,
837
+ "train_steps_per_second": 0.005
838
+ }
839
+ ],
840
+ "logging_steps": 1,
841
+ "max_steps": 117,
842
+ "num_input_tokens_seen": 0,
843
+ "num_train_epochs": 1,
844
+ "save_steps": 100,
845
+ "stateful_callbacks": {
846
+ "TrainerControl": {
847
+ "args": {
848
+ "should_epoch_stop": false,
849
+ "should_evaluate": false,
850
+ "should_log": false,
851
+ "should_save": true,
852
+ "should_training_stop": true
853
+ },
854
+ "attributes": {}
855
+ }
856
+ },
857
+ "total_flos": 609898777804800.0,
858
+ "train_batch_size": 4,
859
+ "trial_name": null,
860
+ "trial_params": null
861
+ }
training_loss.png ADDED