Hugging Face's logo

GaetanMichelet
/
Mistral-7B_task-1_120-samples_config-1

PEFT
TensorBoard
Safetensors
mistral
alignment-handbook
trl
sft
Generated from Trainer
4-bit precision
bitsandbytes
Model card Files Metrics Community
GaetanMichelet commited on
Commit
f0d2c74
•
1 Parent(s): 725b657

Model save

Browse files
Files changed (5) hide show
  1. README.md +74 -0
  2. adapter_model.safetensors +1 -1
  3. all_results.json +9 -0
  4. train_results.json +9 -0
  5. trainer_state.json +523 -0
README.md ADDED
@@ -0,0 +1,74 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: mistralai/Mistral-7B-Instruct-v0.3
3
+ library_name: peft
4
+ license: apache-2.0
5
+ tags:
6
+ - trl
7
+ - sft
8
+ - generated_from_trainer
9
+ model-index:
10
+ - name: Mistral-7B_task-1_120-samples_config-1
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # Mistral-7B_task-1_120-samples_config-1
18
+
19
+ This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.3](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3) on the None dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 2.1311
22
+
23
+ ## Model description
24
+
25
+ More information needed
26
+
27
+ ## Intended uses & limitations
28
+
29
+ More information needed
30
+
31
+ ## Training and evaluation data
32
+
33
+ More information needed
34
+
35
+ ## Training procedure
36
+
37
+ ### Training hyperparameters
38
+
39
+ The following hyperparameters were used during training:
40
+ - learning_rate: 0.0001
41
+ - train_batch_size: 1
42
+ - eval_batch_size: 1
43
+ - seed: 42
44
+ - distributed_type: multi-GPU
45
+ - gradient_accumulation_steps: 8
46
+ - total_train_batch_size: 8
47
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
48
+ - lr_scheduler_type: cosine
49
+ - lr_scheduler_warmup_ratio: 0.1
50
+ - num_epochs: 50
51
+
52
+ ### Training results
53
+
54
+ | Training Loss | Epoch | Step | Validation Loss |
55
+ |:-------------:|:-----:|:----:|:---------------:|
56
+ | 1.8479 | 1.0 | 11 | 1.7767 |
57
+ | 1.3159 | 2.0 | 22 | 1.3060 |
58
+ | 1.0602 | 3.0 | 33 | 1.1514 |
59
+ | 0.7949 | 4.0 | 44 | 1.1724 |
60
+ | 0.5137 | 5.0 | 55 | 1.3468 |
61
+ | 0.2333 | 6.0 | 66 | 1.6914 |
62
+ | 0.1177 | 7.0 | 77 | 1.8391 |
63
+ | 0.0738 | 8.0 | 88 | 1.8831 |
64
+ | 0.0676 | 9.0 | 99 | 1.9784 |
65
+ | 0.0588 | 10.0 | 110 | 2.1311 |
66
+
67
+
68
+ ### Framework versions
69
+
70
+ - PEFT 0.12.0
71
+ - Transformers 4.44.0
72
+ - Pytorch 2.1.2+cu121
73
+ - Datasets 2.20.0
74
+ - Tokenizers 0.19.1
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ad457df76230354db1ae5340326b873f4dcf0ff17b07f854f0d06474d752758c
3
  size 167832240
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9eab5977538eaebfd1bf578b93adc413c0da46ba7a55277eca424e3be60a70ec
3
  size 167832240
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "total_flos": 1.3256024760778752e+16,
4
+ "train_loss": 0.674870581145991,
5
+ "train_runtime": 1036.2269,
6
+ "train_samples": 88,
7
+ "train_samples_per_second": 4.246,
8
+ "train_steps_per_second": 0.531
9
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "total_flos": 1.3256024760778752e+16,
4
+ "train_loss": 0.674870581145991,
5
+ "train_runtime": 1036.2269,
6
+ "train_samples": 88,
7
+ "train_samples_per_second": 4.246,
8
+ "train_steps_per_second": 0.531
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,523 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 1.1513956785202026,
3
+ "best_model_checkpoint": "data/Mistral-7B_task-1_120-samples_config-1/checkpoint-33",
4
+ "epoch": 10.0,
5
+ "eval_steps": 500,
6
+ "global_step": 110,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.09090909090909091,
13
+ "grad_norm": 6.151435375213623,
14
+ "learning_rate": 1.818181818181818e-06,
15
+ "loss": 2.4652,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.18181818181818182,
20
+ "grad_norm": 6.202335834503174,
21
+ "learning_rate": 3.636363636363636e-06,
22
+ "loss": 2.5638,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.36363636363636365,
27
+ "grad_norm": 5.723293304443359,
28
+ "learning_rate": 7.272727272727272e-06,
29
+ "loss": 2.2237,
30
+ "step": 4
31
+ },
32
+ {
33
+ "epoch": 0.5454545454545454,
34
+ "grad_norm": 5.8827128410339355,
35
+ "learning_rate": 1.0909090909090909e-05,
36
+ "loss": 2.1827,
37
+ "step": 6
38
+ },
39
+ {
40
+ "epoch": 0.7272727272727273,
41
+ "grad_norm": 5.391604900360107,
42
+ "learning_rate": 1.4545454545454545e-05,
43
+ "loss": 2.2469,
44
+ "step": 8
45
+ },
46
+ {
47
+ "epoch": 0.9090909090909091,
48
+ "grad_norm": 4.755679130554199,
49
+ "learning_rate": 1.8181818181818182e-05,
50
+ "loss": 1.8479,
51
+ "step": 10
52
+ },
53
+ {
54
+ "epoch": 1.0,
55
+ "eval_loss": 1.7766928672790527,
56
+ "eval_runtime": 9.6489,
57
+ "eval_samples_per_second": 2.487,
58
+ "eval_steps_per_second": 2.487,
59
+ "step": 11
60
+ },
61
+ {
62
+ "epoch": 1.0909090909090908,
63
+ "grad_norm": 3.921355724334717,
64
+ "learning_rate": 2.1818181818181818e-05,
65
+ "loss": 1.7206,
66
+ "step": 12
67
+ },
68
+ {
69
+ "epoch": 1.2727272727272727,
70
+ "grad_norm": 3.4946775436401367,
71
+ "learning_rate": 2.5454545454545454e-05,
72
+ "loss": 1.7337,
73
+ "step": 14
74
+ },
75
+ {
76
+ "epoch": 1.4545454545454546,
77
+ "grad_norm": 2.374727487564087,
78
+ "learning_rate": 2.909090909090909e-05,
79
+ "loss": 1.4858,
80
+ "step": 16
81
+ },
82
+ {
83
+ "epoch": 1.6363636363636362,
84
+ "grad_norm": 2.04058575630188,
85
+ "learning_rate": 3.272727272727273e-05,
86
+ "loss": 1.3494,
87
+ "step": 18
88
+ },
89
+ {
90
+ "epoch": 1.8181818181818183,
91
+ "grad_norm": 2.1711928844451904,
92
+ "learning_rate": 3.6363636363636364e-05,
93
+ "loss": 1.3026,
94
+ "step": 20
95
+ },
96
+ {
97
+ "epoch": 2.0,
98
+ "grad_norm": 1.4772785902023315,
99
+ "learning_rate": 4e-05,
100
+ "loss": 1.3159,
101
+ "step": 22
102
+ },
103
+ {
104
+ "epoch": 2.0,
105
+ "eval_loss": 1.3060346841812134,
106
+ "eval_runtime": 9.6501,
107
+ "eval_samples_per_second": 2.487,
108
+ "eval_steps_per_second": 2.487,
109
+ "step": 22
110
+ },
111
+ {
112
+ "epoch": 2.1818181818181817,
113
+ "grad_norm": 1.4939863681793213,
114
+ "learning_rate": 4.3636363636363636e-05,
115
+ "loss": 1.2751,
116
+ "step": 24
117
+ },
118
+ {
119
+ "epoch": 2.3636363636363638,
120
+ "grad_norm": 1.551141381263733,
121
+ "learning_rate": 4.7272727272727275e-05,
122
+ "loss": 1.251,
123
+ "step": 26
124
+ },
125
+ {
126
+ "epoch": 2.5454545454545454,
127
+ "grad_norm": 1.5137313604354858,
128
+ "learning_rate": 5.090909090909091e-05,
129
+ "loss": 1.0985,
130
+ "step": 28
131
+ },
132
+ {
133
+ "epoch": 2.7272727272727275,
134
+ "grad_norm": 1.3787038326263428,
135
+ "learning_rate": 5.4545454545454546e-05,
136
+ "loss": 1.0334,
137
+ "step": 30
138
+ },
139
+ {
140
+ "epoch": 2.909090909090909,
141
+ "grad_norm": 1.6689972877502441,
142
+ "learning_rate": 5.818181818181818e-05,
143
+ "loss": 1.0602,
144
+ "step": 32
145
+ },
146
+ {
147
+ "epoch": 3.0,
148
+ "eval_loss": 1.1513956785202026,
149
+ "eval_runtime": 9.6608,
150
+ "eval_samples_per_second": 2.484,
151
+ "eval_steps_per_second": 2.484,
152
+ "step": 33
153
+ },
154
+ {
155
+ "epoch": 3.090909090909091,
156
+ "grad_norm": 1.6474758386611938,
157
+ "learning_rate": 6.181818181818182e-05,
158
+ "loss": 1.0549,
159
+ "step": 34
160
+ },
161
+ {
162
+ "epoch": 3.2727272727272725,
163
+ "grad_norm": 1.4157817363739014,
164
+ "learning_rate": 6.545454545454546e-05,
165
+ "loss": 0.9351,
166
+ "step": 36
167
+ },
168
+ {
169
+ "epoch": 3.4545454545454546,
170
+ "grad_norm": 1.760252833366394,
171
+ "learning_rate": 6.90909090909091e-05,
172
+ "loss": 0.9018,
173
+ "step": 38
174
+ },
175
+ {
176
+ "epoch": 3.6363636363636362,
177
+ "grad_norm": 1.7269090414047241,
178
+ "learning_rate": 7.272727272727273e-05,
179
+ "loss": 0.8603,
180
+ "step": 40
181
+ },
182
+ {
183
+ "epoch": 3.8181818181818183,
184
+ "grad_norm": 1.7982224225997925,
185
+ "learning_rate": 7.636363636363637e-05,
186
+ "loss": 0.8153,
187
+ "step": 42
188
+ },
189
+ {
190
+ "epoch": 4.0,
191
+ "grad_norm": 1.8845890760421753,
192
+ "learning_rate": 8e-05,
193
+ "loss": 0.7949,
194
+ "step": 44
195
+ },
196
+ {
197
+ "epoch": 4.0,
198
+ "eval_loss": 1.172354817390442,
199
+ "eval_runtime": 9.6525,
200
+ "eval_samples_per_second": 2.486,
201
+ "eval_steps_per_second": 2.486,
202
+ "step": 44
203
+ },
204
+ {
205
+ "epoch": 4.181818181818182,
206
+ "grad_norm": 1.8724236488342285,
207
+ "learning_rate": 8.363636363636364e-05,
208
+ "loss": 0.6183,
209
+ "step": 46
210
+ },
211
+ {
212
+ "epoch": 4.363636363636363,
213
+ "grad_norm": 2.1722185611724854,
214
+ "learning_rate": 8.727272727272727e-05,
215
+ "loss": 0.5739,
216
+ "step": 48
217
+ },
218
+ {
219
+ "epoch": 4.545454545454545,
220
+ "grad_norm": 2.6997480392456055,
221
+ "learning_rate": 9.090909090909092e-05,
222
+ "loss": 0.5654,
223
+ "step": 50
224
+ },
225
+ {
226
+ "epoch": 4.7272727272727275,
227
+ "grad_norm": 2.968306541442871,
228
+ "learning_rate": 9.454545454545455e-05,
229
+ "loss": 0.5868,
230
+ "step": 52
231
+ },
232
+ {
233
+ "epoch": 4.909090909090909,
234
+ "grad_norm": 2.91367244720459,
235
+ "learning_rate": 9.818181818181818e-05,
236
+ "loss": 0.5137,
237
+ "step": 54
238
+ },
239
+ {
240
+ "epoch": 5.0,
241
+ "eval_loss": 1.346767783164978,
242
+ "eval_runtime": 9.6556,
243
+ "eval_samples_per_second": 2.486,
244
+ "eval_steps_per_second": 2.486,
245
+ "step": 55
246
+ },
247
+ {
248
+ "epoch": 5.090909090909091,
249
+ "grad_norm": 2.089282274246216,
250
+ "learning_rate": 9.999899300364532e-05,
251
+ "loss": 0.421,
252
+ "step": 56
253
+ },
254
+ {
255
+ "epoch": 5.2727272727272725,
256
+ "grad_norm": 2.366018772125244,
257
+ "learning_rate": 9.99909372761763e-05,
258
+ "loss": 0.274,
259
+ "step": 58
260
+ },
261
+ {
262
+ "epoch": 5.454545454545454,
263
+ "grad_norm": 2.2366552352905273,
264
+ "learning_rate": 9.997482711915927e-05,
265
+ "loss": 0.27,
266
+ "step": 60
267
+ },
268
+ {
269
+ "epoch": 5.636363636363637,
270
+ "grad_norm": 2.655911684036255,
271
+ "learning_rate": 9.99506651282272e-05,
272
+ "loss": 0.2622,
273
+ "step": 62
274
+ },
275
+ {
276
+ "epoch": 5.818181818181818,
277
+ "grad_norm": 4.180952548980713,
278
+ "learning_rate": 9.991845519630678e-05,
279
+ "loss": 0.2599,
280
+ "step": 64
281
+ },
282
+ {
283
+ "epoch": 6.0,
284
+ "grad_norm": 3.6061222553253174,
285
+ "learning_rate": 9.987820251299122e-05,
286
+ "loss": 0.2333,
287
+ "step": 66
288
+ },
289
+ {
290
+ "epoch": 6.0,
291
+ "eval_loss": 1.6914259195327759,
292
+ "eval_runtime": 9.6466,
293
+ "eval_samples_per_second": 2.488,
294
+ "eval_steps_per_second": 2.488,
295
+ "step": 66
296
+ },
297
+ {
298
+ "epoch": 6.181818181818182,
299
+ "grad_norm": 2.1230361461639404,
300
+ "learning_rate": 9.982991356370404e-05,
301
+ "loss": 0.1391,
302
+ "step": 68
303
+ },
304
+ {
305
+ "epoch": 6.363636363636363,
306
+ "grad_norm": 3.146448850631714,
307
+ "learning_rate": 9.977359612865423e-05,
308
+ "loss": 0.1079,
309
+ "step": 70
310
+ },
311
+ {
312
+ "epoch": 6.545454545454545,
313
+ "grad_norm": 3.548719644546509,
314
+ "learning_rate": 9.970925928158274e-05,
315
+ "loss": 0.1151,
316
+ "step": 72
317
+ },
318
+ {
319
+ "epoch": 6.7272727272727275,
320
+ "grad_norm": 2.8574233055114746,
321
+ "learning_rate": 9.963691338830044e-05,
322
+ "loss": 0.1403,
323
+ "step": 74
324
+ },
325
+ {
326
+ "epoch": 6.909090909090909,
327
+ "grad_norm": 3.1925230026245117,
328
+ "learning_rate": 9.955657010501806e-05,
329
+ "loss": 0.1177,
330
+ "step": 76
331
+ },
332
+ {
333
+ "epoch": 7.0,
334
+ "eval_loss": 1.8391222953796387,
335
+ "eval_runtime": 9.6552,
336
+ "eval_samples_per_second": 2.486,
337
+ "eval_steps_per_second": 2.486,
338
+ "step": 77
339
+ },
340
+ {
341
+ "epoch": 7.090909090909091,
342
+ "grad_norm": 2.0694875717163086,
343
+ "learning_rate": 9.946824237646824e-05,
344
+ "loss": 0.0844,
345
+ "step": 78
346
+ },
347
+ {
348
+ "epoch": 7.2727272727272725,
349
+ "grad_norm": 2.7770867347717285,
350
+ "learning_rate": 9.937194443381972e-05,
351
+ "loss": 0.0774,
352
+ "step": 80
353
+ },
354
+ {
355
+ "epoch": 7.454545454545454,
356
+ "grad_norm": 1.8506356477737427,
357
+ "learning_rate": 9.926769179238466e-05,
358
+ "loss": 0.0646,
359
+ "step": 82
360
+ },
361
+ {
362
+ "epoch": 7.636363636363637,
363
+ "grad_norm": 2.3127522468566895,
364
+ "learning_rate": 9.915550124911866e-05,
365
+ "loss": 0.0926,
366
+ "step": 84
367
+ },
368
+ {
369
+ "epoch": 7.818181818181818,
370
+ "grad_norm": 1.9697074890136719,
371
+ "learning_rate": 9.903539087991462e-05,
372
+ "loss": 0.0568,
373
+ "step": 86
374
+ },
375
+ {
376
+ "epoch": 8.0,
377
+ "grad_norm": 1.8619533777236938,
378
+ "learning_rate": 9.890738003669029e-05,
379
+ "loss": 0.0738,
380
+ "step": 88
381
+ },
382
+ {
383
+ "epoch": 8.0,
384
+ "eval_loss": 1.8831233978271484,
385
+ "eval_runtime": 9.6565,
386
+ "eval_samples_per_second": 2.485,
387
+ "eval_steps_per_second": 2.485,
388
+ "step": 88
389
+ },
390
+ {
391
+ "epoch": 8.181818181818182,
392
+ "grad_norm": 1.0307751893997192,
393
+ "learning_rate": 9.877148934427037e-05,
394
+ "loss": 0.0405,
395
+ "step": 90
396
+ },
397
+ {
398
+ "epoch": 8.363636363636363,
399
+ "grad_norm": 1.3500251770019531,
400
+ "learning_rate": 9.862774069706346e-05,
401
+ "loss": 0.0493,
402
+ "step": 92
403
+ },
404
+ {
405
+ "epoch": 8.545454545454545,
406
+ "grad_norm": 2.4816551208496094,
407
+ "learning_rate": 9.847615725553456e-05,
408
+ "loss": 0.0387,
409
+ "step": 94
410
+ },
411
+ {
412
+ "epoch": 8.727272727272727,
413
+ "grad_norm": 2.9809985160827637,
414
+ "learning_rate": 9.831676344247342e-05,
415
+ "loss": 0.0429,
416
+ "step": 96
417
+ },
418
+ {
419
+ "epoch": 8.909090909090908,
420
+ "grad_norm": 3.9411723613739014,
421
+ "learning_rate": 9.814958493905963e-05,
422
+ "loss": 0.0676,
423
+ "step": 98
424
+ },
425
+ {
426
+ "epoch": 9.0,
427
+ "eval_loss": 1.9783605337142944,
428
+ "eval_runtime": 9.6493,
429
+ "eval_samples_per_second": 2.487,
430
+ "eval_steps_per_second": 2.487,
431
+ "step": 99
432
+ },
433
+ {
434
+ "epoch": 9.090909090909092,
435
+ "grad_norm": 1.2140812873840332,
436
+ "learning_rate": 9.797464868072488e-05,
437
+ "loss": 0.0389,
438
+ "step": 100
439
+ },
440
+ {
441
+ "epoch": 9.272727272727273,
442
+ "grad_norm": 0.9542932510375977,
443
+ "learning_rate": 9.779198285281325e-05,
444
+ "loss": 0.0228,
445
+ "step": 102
446
+ },
447
+ {
448
+ "epoch": 9.454545454545455,
449
+ "grad_norm": 3.229567289352417,
450
+ "learning_rate": 9.760161688604008e-05,
451
+ "loss": 0.0349,
452
+ "step": 104
453
+ },
454
+ {
455
+ "epoch": 9.636363636363637,
456
+ "grad_norm": 1.3410265445709229,
457
+ "learning_rate": 9.740358145174998e-05,
458
+ "loss": 0.023,
459
+ "step": 106
460
+ },
461
+ {
462
+ "epoch": 9.818181818181818,
463
+ "grad_norm": 2.039607048034668,
464
+ "learning_rate": 9.719790845697533e-05,
465
+ "loss": 0.0478,
466
+ "step": 108
467
+ },
468
+ {
469
+ "epoch": 10.0,
470
+ "grad_norm": 4.932439804077148,
471
+ "learning_rate": 9.698463103929542e-05,
472
+ "loss": 0.0588,
473
+ "step": 110
474
+ },
475
+ {
476
+ "epoch": 10.0,
477
+ "eval_loss": 2.13114857673645,
478
+ "eval_runtime": 9.6508,
479
+ "eval_samples_per_second": 2.487,
480
+ "eval_steps_per_second": 2.487,
481
+ "step": 110
482
+ },
483
+ {
484
+ "epoch": 10.0,
485
+ "step": 110,
486
+ "total_flos": 1.3256024760778752e+16,
487
+ "train_loss": 0.674870581145991,
488
+ "train_runtime": 1036.2269,
489
+ "train_samples_per_second": 4.246,
490
+ "train_steps_per_second": 0.531
491
+ }
492
+ ],
493
+ "logging_steps": 2,
494
+ "max_steps": 550,
495
+ "num_input_tokens_seen": 0,
496
+ "num_train_epochs": 50,
497
+ "save_steps": 25,
498
+ "stateful_callbacks": {
499
+ "EarlyStoppingCallback": {
500
+ "args": {
501
+ "early_stopping_patience": 7,
502
+ "early_stopping_threshold": 0.0
503
+ },
504
+ "attributes": {
505
+ "early_stopping_patience_counter": 0
506
+ }
507
+ },
508
+ "TrainerControl": {
509
+ "args": {
510
+ "should_epoch_stop": false,
511
+ "should_evaluate": false,
512
+ "should_log": false,
513
+ "should_save": true,
514
+ "should_training_stop": true
515
+ },
516
+ "attributes": {}
517
+ }
518
+ },
519
+ "total_flos": 1.3256024760778752e+16,
520
+ "train_batch_size": 1,
521
+ "trial_name": null,
522
+ "trial_params": null
523
+ }