chansung commited on
Commit
ecdd06c
1 Parent(s): 8bb27ed

Model save

Browse files
README.md ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: apache-2.0
4
+ base_model: mistralai/Mistral-7B-v0.3
5
+ tags:
6
+ - trl
7
+ - sft
8
+ - generated_from_trainer
9
+ datasets:
10
+ - generator
11
+ model-index:
12
+ - name: mistral-7b-0.3-gpt4o_100k_summarize-fft
13
+ results: []
14
+ ---
15
+
16
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
+ should probably proofread and complete it, then remove this comment. -->
18
+
19
+ # mistral-7b-0.3-gpt4o_100k_summarize-fft
20
+
21
+ This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on the generator dataset.
22
+ It achieves the following results on the evaluation set:
23
+ - Loss: 7.1395
24
+
25
+ ## Model description
26
+
27
+ More information needed
28
+
29
+ ## Intended uses & limitations
30
+
31
+ More information needed
32
+
33
+ ## Training and evaluation data
34
+
35
+ More information needed
36
+
37
+ ## Training procedure
38
+
39
+ ### Training hyperparameters
40
+
41
+ The following hyperparameters were used during training:
42
+ - learning_rate: 0.0002
43
+ - train_batch_size: 2
44
+ - eval_batch_size: 2
45
+ - seed: 42
46
+ - distributed_type: multi-GPU
47
+ - num_devices: 8
48
+ - gradient_accumulation_steps: 2
49
+ - total_train_batch_size: 32
50
+ - total_eval_batch_size: 16
51
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
52
+ - lr_scheduler_type: cosine
53
+ - lr_scheduler_warmup_ratio: 0.1
54
+ - num_epochs: 1
55
+
56
+ ### Training results
57
+
58
+ | Training Loss | Epoch | Step | Validation Loss |
59
+ |:-------------:|:-----:|:----:|:---------------:|
60
+ | 3.9549 | 1.0 | 438 | 7.1395 |
61
+
62
+
63
+ ### Framework versions
64
+
65
+ - Transformers 4.45.1
66
+ - Pytorch 2.4.1+cu121
67
+ - Datasets 3.0.1
68
+ - Tokenizers 0.20.0
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 22900899840000.0,
4
+ "train_loss": 6.294099572042352,
5
+ "train_runtime": 7431.9326,
6
+ "train_samples": 115376,
7
+ "train_samples_per_second": 1.884,
8
+ "train_steps_per_second": 0.059
9
+ }
generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 32768,
4
+ "eos_token_id": 32769,
5
+ "pad_token_id": 32769,
6
+ "transformers_version": "4.45.1"
7
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 22900899840000.0,
4
+ "train_loss": 6.294099572042352,
5
+ "train_runtime": 7431.9326,
6
+ "train_samples": 115376,
7
+ "train_samples_per_second": 1.884,
8
+ "train_steps_per_second": 0.059
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,666 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 438,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.00228310502283105,
13
+ "grad_norm": 240.46628016222775,
14
+ "learning_rate": 4.5454545454545455e-06,
15
+ "loss": 1.7929,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.01141552511415525,
20
+ "grad_norm": 245.40050686537415,
21
+ "learning_rate": 2.272727272727273e-05,
22
+ "loss": 2.5951,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.0228310502283105,
27
+ "grad_norm": 108.5355771736147,
28
+ "learning_rate": 4.545454545454546e-05,
29
+ "loss": 2.3381,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 0.03424657534246575,
34
+ "grad_norm": 92.10609311564825,
35
+ "learning_rate": 6.818181818181818e-05,
36
+ "loss": 1.9168,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.045662100456621,
41
+ "grad_norm": 1166.1055385039233,
42
+ "learning_rate": 9.090909090909092e-05,
43
+ "loss": 13.6037,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 0.05707762557077625,
48
+ "grad_norm": 695.5402369585504,
49
+ "learning_rate": 0.00011363636363636365,
50
+ "loss": 13.6584,
51
+ "step": 25
52
+ },
53
+ {
54
+ "epoch": 0.0684931506849315,
55
+ "grad_norm": 761.1239739280335,
56
+ "learning_rate": 0.00013636363636363637,
57
+ "loss": 8.4136,
58
+ "step": 30
59
+ },
60
+ {
61
+ "epoch": 0.07990867579908675,
62
+ "grad_norm": 1578.3027737091188,
63
+ "learning_rate": 0.0001590909090909091,
64
+ "loss": 8.6665,
65
+ "step": 35
66
+ },
67
+ {
68
+ "epoch": 0.091324200913242,
69
+ "grad_norm": 65.2902593433997,
70
+ "learning_rate": 0.00018181818181818183,
71
+ "loss": 9.9526,
72
+ "step": 40
73
+ },
74
+ {
75
+ "epoch": 0.10273972602739725,
76
+ "grad_norm": 92.74211625664995,
77
+ "learning_rate": 0.00019999682111362368,
78
+ "loss": 7.7814,
79
+ "step": 45
80
+ },
81
+ {
82
+ "epoch": 0.1141552511415525,
83
+ "grad_norm": 388.8492851786365,
84
+ "learning_rate": 0.00019988558131018186,
85
+ "loss": 7.643,
86
+ "step": 50
87
+ },
88
+ {
89
+ "epoch": 0.12557077625570776,
90
+ "grad_norm": 52.42068791324127,
91
+ "learning_rate": 0.0001996155992365444,
92
+ "loss": 10.8328,
93
+ "step": 55
94
+ },
95
+ {
96
+ "epoch": 0.136986301369863,
97
+ "grad_norm": 67.22123396518563,
98
+ "learning_rate": 0.00019918730395931649,
99
+ "loss": 12.2142,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 0.14840182648401826,
104
+ "grad_norm": 108.11322168523512,
105
+ "learning_rate": 0.00019860137614295168,
106
+ "loss": 10.6018,
107
+ "step": 65
108
+ },
109
+ {
110
+ "epoch": 0.1598173515981735,
111
+ "grad_norm": 840.8723407449802,
112
+ "learning_rate": 0.00019785874696801202,
113
+ "loss": 11.7075,
114
+ "step": 70
115
+ },
116
+ {
117
+ "epoch": 0.17123287671232876,
118
+ "grad_norm": 920.9036553566684,
119
+ "learning_rate": 0.0001969605966512975,
120
+ "loss": 10.8955,
121
+ "step": 75
122
+ },
123
+ {
124
+ "epoch": 0.182648401826484,
125
+ "grad_norm": 877.4272582883378,
126
+ "learning_rate": 0.00019590835257019714,
127
+ "loss": 12.7904,
128
+ "step": 80
129
+ },
130
+ {
131
+ "epoch": 0.19406392694063926,
132
+ "grad_norm": 912.8271704562273,
133
+ "learning_rate": 0.00019470368699424218,
134
+ "loss": 10.7954,
135
+ "step": 85
136
+ },
137
+ {
138
+ "epoch": 0.2054794520547945,
139
+ "grad_norm": 1008.5135637969497,
140
+ "learning_rate": 0.00019334851442746664,
141
+ "loss": 11.2705,
142
+ "step": 90
143
+ },
144
+ {
145
+ "epoch": 0.21689497716894976,
146
+ "grad_norm": 38.472345652966226,
147
+ "learning_rate": 0.00019184498856579868,
148
+ "loss": 8.4385,
149
+ "step": 95
150
+ },
151
+ {
152
+ "epoch": 0.228310502283105,
153
+ "grad_norm": 57.1422911131311,
154
+ "learning_rate": 0.00019019549887431877,
155
+ "loss": 9.6665,
156
+ "step": 100
157
+ },
158
+ {
159
+ "epoch": 0.23972602739726026,
160
+ "grad_norm": 88.32108375036853,
161
+ "learning_rate": 0.00018840266678982342,
162
+ "loss": 7.5442,
163
+ "step": 105
164
+ },
165
+ {
166
+ "epoch": 0.2511415525114155,
167
+ "grad_norm": 347.05998093496095,
168
+ "learning_rate": 0.00018646934155473022,
169
+ "loss": 7.2958,
170
+ "step": 110
171
+ },
172
+ {
173
+ "epoch": 0.2625570776255708,
174
+ "grad_norm": 20.572286439019486,
175
+ "learning_rate": 0.00018439859568894463,
176
+ "loss": 7.3386,
177
+ "step": 115
178
+ },
179
+ {
180
+ "epoch": 0.273972602739726,
181
+ "grad_norm": 38.351388562103665,
182
+ "learning_rate": 0.00018219372010688515,
183
+ "loss": 8.012,
184
+ "step": 120
185
+ },
186
+ {
187
+ "epoch": 0.2853881278538813,
188
+ "grad_norm": 37.37105945079517,
189
+ "learning_rate": 0.00017985821888742685,
190
+ "loss": 7.1013,
191
+ "step": 125
192
+ },
193
+ {
194
+ "epoch": 0.2968036529680365,
195
+ "grad_norm": 26.7507850568603,
196
+ "learning_rate": 0.00017739580370507532,
197
+ "loss": 6.882,
198
+ "step": 130
199
+ },
200
+ {
201
+ "epoch": 0.3082191780821918,
202
+ "grad_norm": 88.7335799340544,
203
+ "learning_rate": 0.00017481038793122088,
204
+ "loss": 6.7721,
205
+ "step": 135
206
+ },
207
+ {
208
+ "epoch": 0.319634703196347,
209
+ "grad_norm": 22.97956758710175,
210
+ "learning_rate": 0.0001721060804148482,
211
+ "loss": 6.9621,
212
+ "step": 140
213
+ },
214
+ {
215
+ "epoch": 0.3310502283105023,
216
+ "grad_norm": 20.8606176285665,
217
+ "learning_rate": 0.00016928717895258437,
218
+ "loss": 6.7809,
219
+ "step": 145
220
+ },
221
+ {
222
+ "epoch": 0.3424657534246575,
223
+ "grad_norm": 26.219570138783702,
224
+ "learning_rate": 0.0001663581634584641,
225
+ "loss": 6.9626,
226
+ "step": 150
227
+ },
228
+ {
229
+ "epoch": 0.3538812785388128,
230
+ "grad_norm": 167.86360436359567,
231
+ "learning_rate": 0.00016332368884426626,
232
+ "loss": 7.6471,
233
+ "step": 155
234
+ },
235
+ {
236
+ "epoch": 0.365296803652968,
237
+ "grad_norm": 430.63515082965046,
238
+ "learning_rate": 0.0001601885776217367,
239
+ "loss": 7.1777,
240
+ "step": 160
241
+ },
242
+ {
243
+ "epoch": 0.3767123287671233,
244
+ "grad_norm": 25.678714469291474,
245
+ "learning_rate": 0.00015695781223845441,
246
+ "loss": 6.9389,
247
+ "step": 165
248
+ },
249
+ {
250
+ "epoch": 0.3881278538812785,
251
+ "grad_norm": 180.18708109632934,
252
+ "learning_rate": 0.0001536365271595212,
253
+ "loss": 7.4995,
254
+ "step": 170
255
+ },
256
+ {
257
+ "epoch": 0.3995433789954338,
258
+ "grad_norm": 50.63109221066056,
259
+ "learning_rate": 0.00015023000070765884,
260
+ "loss": 6.6471,
261
+ "step": 175
262
+ },
263
+ {
264
+ "epoch": 0.410958904109589,
265
+ "grad_norm": 138.11974530731086,
266
+ "learning_rate": 0.0001467436466746814,
267
+ "loss": 6.4638,
268
+ "step": 180
269
+ },
270
+ {
271
+ "epoch": 0.4223744292237443,
272
+ "grad_norm": 24.557862163981767,
273
+ "learning_rate": 0.00014318300571767513,
274
+ "loss": 6.4258,
275
+ "step": 185
276
+ },
277
+ {
278
+ "epoch": 0.4337899543378995,
279
+ "grad_norm": 23.71602743273707,
280
+ "learning_rate": 0.0001395537365535585,
281
+ "loss": 6.2647,
282
+ "step": 190
283
+ },
284
+ {
285
+ "epoch": 0.4452054794520548,
286
+ "grad_norm": 40.86377604494575,
287
+ "learning_rate": 0.00013586160696601665,
288
+ "loss": 6.2592,
289
+ "step": 195
290
+ },
291
+ {
292
+ "epoch": 0.45662100456621,
293
+ "grad_norm": 40.35100848104416,
294
+ "learning_rate": 0.00013211248463910262,
295
+ "loss": 6.1296,
296
+ "step": 200
297
+ },
298
+ {
299
+ "epoch": 0.4680365296803653,
300
+ "grad_norm": 8.53161320251149,
301
+ "learning_rate": 0.00012831232783207277,
302
+ "loss": 5.9919,
303
+ "step": 205
304
+ },
305
+ {
306
+ "epoch": 0.4794520547945205,
307
+ "grad_norm": 53.85169619244609,
308
+ "learning_rate": 0.00012446717591027624,
309
+ "loss": 6.1196,
310
+ "step": 210
311
+ },
312
+ {
313
+ "epoch": 0.4908675799086758,
314
+ "grad_norm": 18.702516339707657,
315
+ "learning_rate": 0.00012058313974714746,
316
+ "loss": 6.217,
317
+ "step": 215
318
+ },
319
+ {
320
+ "epoch": 0.502283105022831,
321
+ "grad_norm": 23.340868803327382,
322
+ "learning_rate": 0.00011666639201255506,
323
+ "loss": 5.9684,
324
+ "step": 220
325
+ },
326
+ {
327
+ "epoch": 0.5136986301369864,
328
+ "grad_norm": 27.848028057442086,
329
+ "learning_rate": 0.00011272315736294108,
330
+ "loss": 5.885,
331
+ "step": 225
332
+ },
333
+ {
334
+ "epoch": 0.5251141552511416,
335
+ "grad_norm": 11.302150338648437,
336
+ "learning_rate": 0.0001087597025488413,
337
+ "loss": 5.8254,
338
+ "step": 230
339
+ },
340
+ {
341
+ "epoch": 0.5365296803652968,
342
+ "grad_norm": 7.856659467347087,
343
+ "learning_rate": 0.00010478232645550782,
344
+ "loss": 5.797,
345
+ "step": 235
346
+ },
347
+ {
348
+ "epoch": 0.547945205479452,
349
+ "grad_norm": 17.570218309451846,
350
+ "learning_rate": 0.00010079735009246167,
351
+ "loss": 5.6966,
352
+ "step": 240
353
+ },
354
+ {
355
+ "epoch": 0.5593607305936074,
356
+ "grad_norm": 12.285913129547184,
357
+ "learning_rate": 9.681110654788482e-05,
358
+ "loss": 5.5947,
359
+ "step": 245
360
+ },
361
+ {
362
+ "epoch": 0.5707762557077626,
363
+ "grad_norm": 12.189890918080446,
364
+ "learning_rate": 9.282993092381625e-05,
365
+ "loss": 5.456,
366
+ "step": 250
367
+ },
368
+ {
369
+ "epoch": 0.5821917808219178,
370
+ "grad_norm": 13.447394220420884,
371
+ "learning_rate": 8.886015026814736e-05,
372
+ "loss": 5.3872,
373
+ "step": 255
374
+ },
375
+ {
376
+ "epoch": 0.593607305936073,
377
+ "grad_norm": 8.087645921511065,
378
+ "learning_rate": 8.490807351941753e-05,
379
+ "loss": 5.3234,
380
+ "step": 260
381
+ },
382
+ {
383
+ "epoch": 0.6050228310502284,
384
+ "grad_norm": 6.294622158332865,
385
+ "learning_rate": 8.097998148038985e-05,
386
+ "loss": 5.2182,
387
+ "step": 265
388
+ },
389
+ {
390
+ "epoch": 0.6164383561643836,
391
+ "grad_norm": 6.8880132676177155,
392
+ "learning_rate": 7.708211683634112e-05,
393
+ "loss": 5.1346,
394
+ "step": 270
395
+ },
396
+ {
397
+ "epoch": 0.6278538812785388,
398
+ "grad_norm": 4.3063046925628665,
399
+ "learning_rate": 7.322067423393002e-05,
400
+ "loss": 5.0977,
401
+ "step": 275
402
+ },
403
+ {
404
+ "epoch": 0.639269406392694,
405
+ "grad_norm": 6.202488868271211,
406
+ "learning_rate": 6.940179043641005e-05,
407
+ "loss": 4.9949,
408
+ "step": 280
409
+ },
410
+ {
411
+ "epoch": 0.6506849315068494,
412
+ "grad_norm": 9.748600106096807,
413
+ "learning_rate": 6.563153457083315e-05,
414
+ "loss": 5.0211,
415
+ "step": 285
416
+ },
417
+ {
418
+ "epoch": 0.6621004566210046,
419
+ "grad_norm": 4.021422440954996,
420
+ "learning_rate": 6.191589848274368e-05,
421
+ "loss": 4.9403,
422
+ "step": 290
423
+ },
424
+ {
425
+ "epoch": 0.6735159817351598,
426
+ "grad_norm": 5.203478173074032,
427
+ "learning_rate": 5.82607872136913e-05,
428
+ "loss": 4.8661,
429
+ "step": 295
430
+ },
431
+ {
432
+ "epoch": 0.684931506849315,
433
+ "grad_norm": 7.795245351250236,
434
+ "learning_rate": 5.467200961669619e-05,
435
+ "loss": 4.8218,
436
+ "step": 300
437
+ },
438
+ {
439
+ "epoch": 0.6963470319634704,
440
+ "grad_norm": 6.507211128414033,
441
+ "learning_rate": 5.115526912458113e-05,
442
+ "loss": 4.7079,
443
+ "step": 305
444
+ },
445
+ {
446
+ "epoch": 0.7077625570776256,
447
+ "grad_norm": 3.415005954143531,
448
+ "learning_rate": 4.7716154685841944e-05,
449
+ "loss": 4.7262,
450
+ "step": 310
451
+ },
452
+ {
453
+ "epoch": 0.7191780821917808,
454
+ "grad_norm": 4.701959725615529,
455
+ "learning_rate": 4.4360131882460555e-05,
456
+ "loss": 4.6772,
457
+ "step": 315
458
+ },
459
+ {
460
+ "epoch": 0.730593607305936,
461
+ "grad_norm": 7.243992416761161,
462
+ "learning_rate": 4.109253424377772e-05,
463
+ "loss": 4.6001,
464
+ "step": 320
465
+ },
466
+ {
467
+ "epoch": 0.7420091324200914,
468
+ "grad_norm": 5.597934572299491,
469
+ "learning_rate": 3.791855477022903e-05,
470
+ "loss": 4.5783,
471
+ "step": 325
472
+ },
473
+ {
474
+ "epoch": 0.7534246575342466,
475
+ "grad_norm": 7.0945462547518545,
476
+ "learning_rate": 3.4843237680415156e-05,
477
+ "loss": 4.5086,
478
+ "step": 330
479
+ },
480
+ {
481
+ "epoch": 0.7648401826484018,
482
+ "grad_norm": 5.40993943834967,
483
+ "learning_rate": 3.1871470394622404e-05,
484
+ "loss": 4.396,
485
+ "step": 335
486
+ },
487
+ {
488
+ "epoch": 0.776255707762557,
489
+ "grad_norm": 4.514795217834535,
490
+ "learning_rate": 2.9007975767533714e-05,
491
+ "loss": 4.4116,
492
+ "step": 340
493
+ },
494
+ {
495
+ "epoch": 0.7876712328767124,
496
+ "grad_norm": 6.619199704533781,
497
+ "learning_rate": 2.625730458247362e-05,
498
+ "loss": 4.3946,
499
+ "step": 345
500
+ },
501
+ {
502
+ "epoch": 0.7990867579908676,
503
+ "grad_norm": 6.797118824330897,
504
+ "learning_rate": 2.3623828319116748e-05,
505
+ "loss": 4.3655,
506
+ "step": 350
507
+ },
508
+ {
509
+ "epoch": 0.8105022831050228,
510
+ "grad_norm": 6.319532987233403,
511
+ "learning_rate": 2.1111732206152424e-05,
512
+ "loss": 4.3162,
513
+ "step": 355
514
+ },
515
+ {
516
+ "epoch": 0.821917808219178,
517
+ "grad_norm": 3.6539166546644366,
518
+ "learning_rate": 1.8725008569947365e-05,
519
+ "loss": 4.2938,
520
+ "step": 360
521
+ },
522
+ {
523
+ "epoch": 0.8333333333333334,
524
+ "grad_norm": 4.300626146767242,
525
+ "learning_rate": 1.6467450489776582e-05,
526
+ "loss": 4.1979,
527
+ "step": 365
528
+ },
529
+ {
530
+ "epoch": 0.8447488584474886,
531
+ "grad_norm": 4.330204640440713,
532
+ "learning_rate": 1.4342645769705977e-05,
533
+ "loss": 4.1682,
534
+ "step": 370
535
+ },
536
+ {
537
+ "epoch": 0.8561643835616438,
538
+ "grad_norm": 3.3916542716906064,
539
+ "learning_rate": 1.2353971236706564e-05,
540
+ "loss": 4.1933,
541
+ "step": 375
542
+ },
543
+ {
544
+ "epoch": 0.867579908675799,
545
+ "grad_norm": 3.4156576312872726,
546
+ "learning_rate": 1.0504587374062391e-05,
547
+ "loss": 4.0807,
548
+ "step": 380
549
+ },
550
+ {
551
+ "epoch": 0.8789954337899544,
552
+ "grad_norm": 3.740969173497314,
553
+ "learning_rate": 8.797433298600622e-06,
554
+ "loss": 4.0922,
555
+ "step": 385
556
+ },
557
+ {
558
+ "epoch": 0.8904109589041096,
559
+ "grad_norm": 3.4358795428140096,
560
+ "learning_rate": 7.235222089726279e-06,
561
+ "loss": 4.129,
562
+ "step": 390
563
+ },
564
+ {
565
+ "epoch": 0.9018264840182648,
566
+ "grad_norm": 4.4709463768242,
567
+ "learning_rate": 5.82043647768502e-06,
568
+ "loss": 4.0451,
569
+ "step": 395
570
+ },
571
+ {
572
+ "epoch": 0.91324200913242,
573
+ "grad_norm": 3.239079694881163,
574
+ "learning_rate": 4.555324897906132e-06,
575
+ "loss": 4.0609,
576
+ "step": 400
577
+ },
578
+ {
579
+ "epoch": 0.9246575342465754,
580
+ "grad_norm": 4.365463466537055,
581
+ "learning_rate": 3.441897917696679e-06,
582
+ "loss": 4.0053,
583
+ "step": 405
584
+ },
585
+ {
586
+ "epoch": 0.9360730593607306,
587
+ "grad_norm": 3.4103467721542082,
588
+ "learning_rate": 2.4819250409651607e-06,
589
+ "loss": 4.0147,
590
+ "step": 410
591
+ },
592
+ {
593
+ "epoch": 0.9474885844748858,
594
+ "grad_norm": 2.424364857956352,
595
+ "learning_rate": 1.6769318960533464e-06,
596
+ "loss": 3.9732,
597
+ "step": 415
598
+ },
599
+ {
600
+ "epoch": 0.958904109589041,
601
+ "grad_norm": 2.681966720958818,
602
+ "learning_rate": 1.0281978111449375e-06,
603
+ "loss": 4.0434,
604
+ "step": 420
605
+ },
606
+ {
607
+ "epoch": 0.9703196347031964,
608
+ "grad_norm": 2.756086031182227,
609
+ "learning_rate": 5.367537811046485e-07,
610
+ "loss": 4.028,
611
+ "step": 425
612
+ },
613
+ {
614
+ "epoch": 0.9817351598173516,
615
+ "grad_norm": 2.613500426580146,
616
+ "learning_rate": 2.0338082897886079e-07,
617
+ "loss": 3.9626,
618
+ "step": 430
619
+ },
620
+ {
621
+ "epoch": 0.9931506849315068,
622
+ "grad_norm": 2.209338347291564,
623
+ "learning_rate": 2.8608764761639538e-08,
624
+ "loss": 3.9549,
625
+ "step": 435
626
+ },
627
+ {
628
+ "epoch": 1.0,
629
+ "eval_loss": 7.139484405517578,
630
+ "eval_runtime": 2.4665,
631
+ "eval_samples_per_second": 4.865,
632
+ "eval_steps_per_second": 0.405,
633
+ "step": 438
634
+ },
635
+ {
636
+ "epoch": 1.0,
637
+ "step": 438,
638
+ "total_flos": 22900899840000.0,
639
+ "train_loss": 6.294099572042352,
640
+ "train_runtime": 7431.9326,
641
+ "train_samples_per_second": 1.884,
642
+ "train_steps_per_second": 0.059
643
+ }
644
+ ],
645
+ "logging_steps": 5,
646
+ "max_steps": 438,
647
+ "num_input_tokens_seen": 0,
648
+ "num_train_epochs": 1,
649
+ "save_steps": 100,
650
+ "stateful_callbacks": {
651
+ "TrainerControl": {
652
+ "args": {
653
+ "should_epoch_stop": false,
654
+ "should_evaluate": false,
655
+ "should_log": false,
656
+ "should_save": true,
657
+ "should_training_stop": true
658
+ },
659
+ "attributes": {}
660
+ }
661
+ },
662
+ "total_flos": 22900899840000.0,
663
+ "train_batch_size": 2,
664
+ "trial_name": null,
665
+ "trial_params": null
666
+ }