kykim0 commited on
Commit
000fc5c
1 Parent(s): 2ae246d

Model save

Browse files
README.md ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: llama2
4
+ base_model: codellama/CodeLlama-7b-Instruct-hf
5
+ tags:
6
+ - trl
7
+ - sft
8
+ - generated_from_trainer
9
+ datasets:
10
+ - generator
11
+ model-index:
12
+ - name: codellama-7b-ultrachat
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
+ # codellama-7b-ultrachat
20
+
21
+ This model is a fine-tuned version of [codellama/CodeLlama-7b-Instruct-hf](https://huggingface.co/codellama/CodeLlama-7b-Instruct-hf) on the generator dataset.
22
+ It achieves the following results on the evaluation set:
23
+ - Loss: 1.0320
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: 2e-05
43
+ - train_batch_size: 1
44
+ - eval_batch_size: 4
45
+ - seed: 42
46
+ - distributed_type: multi-GPU
47
+ - num_devices: 4
48
+ - gradient_accumulation_steps: 32
49
+ - total_train_batch_size: 128
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: 2
55
+
56
+ ### Training results
57
+
58
+ | Training Loss | Epoch | Step | Validation Loss |
59
+ |:-------------:|:------:|:----:|:---------------:|
60
+ | 1.0317 | 0.9995 | 555 | 1.0397 |
61
+ | 0.9698 | 1.9991 | 1110 | 1.0320 |
62
+
63
+
64
+ ### Framework versions
65
+
66
+ - Transformers 4.44.2
67
+ - Pytorch 2.1.2
68
+ - Datasets 2.14.6
69
+ - Tokenizers 0.19.1
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.9990995047276003,
3
+ "total_flos": 929619498762240.0,
4
+ "train_loss": 1.0262996888375497,
5
+ "train_runtime": 54210.2289,
6
+ "train_samples": 207864,
7
+ "train_samples_per_second": 2.622,
8
+ "train_steps_per_second": 0.02
9
+ }
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "transformers_version": "4.44.2"
6
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.9990995047276003,
3
+ "total_flos": 929619498762240.0,
4
+ "train_loss": 1.0262996888375497,
5
+ "train_runtime": 54210.2289,
6
+ "train_samples": 207864,
7
+ "train_samples_per_second": 2.622,
8
+ "train_steps_per_second": 0.02
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,835 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.9990995047276003,
5
+ "eval_steps": 500,
6
+ "global_step": 1110,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.018009905447996397,
13
+ "grad_norm": 3.8927660727308404,
14
+ "learning_rate": 1.801801801801802e-06,
15
+ "loss": 1.3686,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.03601981089599279,
20
+ "grad_norm": 1.2914340039825571,
21
+ "learning_rate": 3.603603603603604e-06,
22
+ "loss": 1.2618,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.0540297163439892,
27
+ "grad_norm": 0.5117572348414451,
28
+ "learning_rate": 5.405405405405406e-06,
29
+ "loss": 1.1941,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.07203962179198559,
34
+ "grad_norm": 0.3415375067191979,
35
+ "learning_rate": 7.207207207207208e-06,
36
+ "loss": 1.157,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.090049527239982,
41
+ "grad_norm": 0.3050099720948823,
42
+ "learning_rate": 9.00900900900901e-06,
43
+ "loss": 1.1486,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.1080594326879784,
48
+ "grad_norm": 0.2628693083106922,
49
+ "learning_rate": 1.0810810810810812e-05,
50
+ "loss": 1.12,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.1260693381359748,
55
+ "grad_norm": 0.23606513506788576,
56
+ "learning_rate": 1.2612612612612613e-05,
57
+ "loss": 1.1014,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.14407924358397117,
62
+ "grad_norm": 0.24000278532244093,
63
+ "learning_rate": 1.4414414414414416e-05,
64
+ "loss": 1.1087,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.16208914903196758,
69
+ "grad_norm": 0.21377676080662794,
70
+ "learning_rate": 1.6216216216216218e-05,
71
+ "loss": 1.0764,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.180099054479964,
76
+ "grad_norm": 0.1794631392027073,
77
+ "learning_rate": 1.801801801801802e-05,
78
+ "loss": 1.0757,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.19810895992796038,
83
+ "grad_norm": 0.19497374657829,
84
+ "learning_rate": 1.981981981981982e-05,
85
+ "loss": 1.0904,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.2161188653759568,
90
+ "grad_norm": 0.21533071811896856,
91
+ "learning_rate": 1.999599507118322e-05,
92
+ "loss": 1.0765,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.23412877082395317,
97
+ "grad_norm": 0.19307788030973838,
98
+ "learning_rate": 1.998215499120161e-05,
99
+ "loss": 1.0702,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.2521386762719496,
104
+ "grad_norm": 0.18836481163407426,
105
+ "learning_rate": 1.99584439990734e-05,
106
+ "loss": 1.0766,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.270148581719946,
111
+ "grad_norm": 0.21511735322364797,
112
+ "learning_rate": 1.992488554155135e-05,
113
+ "loss": 1.0811,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.28815848716794235,
118
+ "grad_norm": 0.195309434953766,
119
+ "learning_rate": 1.9881512803111797e-05,
120
+ "loss": 1.0658,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.30616839261593876,
125
+ "grad_norm": 0.2421941539327756,
126
+ "learning_rate": 1.9828368673139947e-05,
127
+ "loss": 1.0672,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.32417829806393517,
132
+ "grad_norm": 0.1860112417998846,
133
+ "learning_rate": 1.9765505703518494e-05,
134
+ "loss": 1.0611,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.3421882035119316,
139
+ "grad_norm": 0.17818304810152363,
140
+ "learning_rate": 1.9692986056661355e-05,
141
+ "loss": 1.066,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.360198108959928,
146
+ "grad_norm": 0.2397415845301638,
147
+ "learning_rate": 1.961088144404403e-05,
148
+ "loss": 1.0634,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.37820801440792434,
153
+ "grad_norm": 0.18967589505610974,
154
+ "learning_rate": 1.9519273055291266e-05,
155
+ "loss": 1.0613,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.39621791985592075,
160
+ "grad_norm": 0.2130198103194759,
161
+ "learning_rate": 1.941825147789225e-05,
162
+ "loss": 1.0668,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.41422782530391716,
167
+ "grad_norm": 0.18395677854198192,
168
+ "learning_rate": 1.930791660762262e-05,
169
+ "loss": 1.0619,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.4322377307519136,
174
+ "grad_norm": 0.1865388893667488,
175
+ "learning_rate": 1.9188377549761962e-05,
176
+ "loss": 1.0636,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.45024763619990993,
181
+ "grad_norm": 0.18776091331352887,
182
+ "learning_rate": 1.90597525112044e-05,
183
+ "loss": 1.0514,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.46825754164790634,
188
+ "grad_norm": 0.18243911298468143,
189
+ "learning_rate": 1.8922168683569038e-05,
190
+ "loss": 1.0519,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.48626744709590275,
195
+ "grad_norm": 0.18797424109818742,
196
+ "learning_rate": 1.8775762117425777e-05,
197
+ "loss": 1.0544,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.5042773525438992,
202
+ "grad_norm": 0.18906571186079896,
203
+ "learning_rate": 1.8620677587760916e-05,
204
+ "loss": 1.0567,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.5222872579918956,
209
+ "grad_norm": 0.18393014193254023,
210
+ "learning_rate": 1.8457068450815562e-05,
211
+ "loss": 1.0567,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.540297163439892,
216
+ "grad_norm": 0.19328266852176415,
217
+ "learning_rate": 1.8285096492438424e-05,
218
+ "loss": 1.055,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.5583070688878884,
223
+ "grad_norm": 0.18341259531456078,
224
+ "learning_rate": 1.810493176810292e-05,
225
+ "loss": 1.0537,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.5763169743358847,
230
+ "grad_norm": 0.19070534973528358,
231
+ "learning_rate": 1.7916752434746856e-05,
232
+ "loss": 1.0388,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.5943268797838811,
237
+ "grad_norm": 0.18919026633766578,
238
+ "learning_rate": 1.7720744574600865e-05,
239
+ "loss": 1.054,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.6123367852318775,
244
+ "grad_norm": 0.2324366480100927,
245
+ "learning_rate": 1.7517102011179935e-05,
246
+ "loss": 1.0521,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.6303466906798739,
251
+ "grad_norm": 0.18954585112855193,
252
+ "learning_rate": 1.730602611761989e-05,
253
+ "loss": 1.0442,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.6483565961278703,
258
+ "grad_norm": 0.19841527530500794,
259
+ "learning_rate": 1.7087725617548385e-05,
260
+ "loss": 1.0461,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.6663665015758667,
265
+ "grad_norm": 0.18229995357270518,
266
+ "learning_rate": 1.686241637868734e-05,
267
+ "loss": 1.0416,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.6843764070238632,
272
+ "grad_norm": 0.18303392694652232,
273
+ "learning_rate": 1.6630321199390868e-05,
274
+ "loss": 1.0444,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.7023863124718596,
279
+ "grad_norm": 0.18252570663629492,
280
+ "learning_rate": 1.639166958832985e-05,
281
+ "loss": 1.0382,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.720396217919856,
286
+ "grad_norm": 0.1920090778916924,
287
+ "learning_rate": 1.6146697537540926e-05,
288
+ "loss": 1.0451,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.7384061233678523,
293
+ "grad_norm": 0.182014948065573,
294
+ "learning_rate": 1.5895647289064395e-05,
295
+ "loss": 1.0459,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.7564160288158487,
300
+ "grad_norm": 0.18069534485817443,
301
+ "learning_rate": 1.5638767095401778e-05,
302
+ "loss": 1.0414,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.7744259342638451,
307
+ "grad_norm": 0.1970419139073729,
308
+ "learning_rate": 1.5376310974029872e-05,
309
+ "loss": 1.0379,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.7924358397118415,
314
+ "grad_norm": 0.1831244876657377,
315
+ "learning_rate": 1.5108538456214088e-05,
316
+ "loss": 1.038,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 0.8104457451598379,
321
+ "grad_norm": 0.18766104093935895,
322
+ "learning_rate": 1.4835714330369445e-05,
323
+ "loss": 1.0396,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 0.8284556506078343,
328
+ "grad_norm": 0.20205755605640713,
329
+ "learning_rate": 1.4558108380223013e-05,
330
+ "loss": 1.0312,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 0.8464655560558307,
335
+ "grad_norm": 0.1958644336406297,
336
+ "learning_rate": 1.4275995118036694e-05,
337
+ "loss": 1.0333,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 0.8644754615038271,
342
+ "grad_norm": 0.17729339680552028,
343
+ "learning_rate": 1.3989653513154165e-05,
344
+ "loss": 1.0429,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 0.8824853669518236,
349
+ "grad_norm": 0.17561113142823978,
350
+ "learning_rate": 1.3699366716140434e-05,
351
+ "loss": 1.0419,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 0.9004952723998199,
356
+ "grad_norm": 0.18385449880212262,
357
+ "learning_rate": 1.3405421778786738e-05,
358
+ "loss": 1.0482,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 0.9185051778478163,
363
+ "grad_norm": 0.18110333089233704,
364
+ "learning_rate": 1.3108109370257714e-05,
365
+ "loss": 1.0539,
366
+ "step": 510
367
+ },
368
+ {
369
+ "epoch": 0.9365150832958127,
370
+ "grad_norm": 0.18232534077561582,
371
+ "learning_rate": 1.2807723489661497e-05,
372
+ "loss": 1.0371,
373
+ "step": 520
374
+ },
375
+ {
376
+ "epoch": 0.9545249887438091,
377
+ "grad_norm": 0.17118485590371502,
378
+ "learning_rate": 1.2504561175326986e-05,
379
+ "loss": 1.034,
380
+ "step": 530
381
+ },
382
+ {
383
+ "epoch": 0.9725348941918055,
384
+ "grad_norm": 0.17973311103034112,
385
+ "learning_rate": 1.2198922211075779e-05,
386
+ "loss": 1.0374,
387
+ "step": 540
388
+ },
389
+ {
390
+ "epoch": 0.9905447996398019,
391
+ "grad_norm": 0.18714581161352067,
392
+ "learning_rate": 1.1891108829779166e-05,
393
+ "loss": 1.0317,
394
+ "step": 550
395
+ },
396
+ {
397
+ "epoch": 0.9995497523638001,
398
+ "eval_loss": 1.039684534072876,
399
+ "eval_runtime": 662.247,
400
+ "eval_samples_per_second": 11.876,
401
+ "eval_steps_per_second": 0.743,
402
+ "step": 555
403
+ },
404
+ {
405
+ "epoch": 1.0085547050877983,
406
+ "grad_norm": 0.20548749168254346,
407
+ "learning_rate": 1.158142541449341e-05,
408
+ "loss": 1.0109,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 1.0265646105357946,
413
+ "grad_norm": 0.201629063500619,
414
+ "learning_rate": 1.1270178197468788e-05,
415
+ "loss": 0.9851,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 1.0445745159837911,
420
+ "grad_norm": 0.19089253248727517,
421
+ "learning_rate": 1.0957674957330043e-05,
422
+ "loss": 0.9736,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 1.0625844214317874,
427
+ "grad_norm": 0.18420339461964708,
428
+ "learning_rate": 1.0644224714727683e-05,
429
+ "loss": 0.9918,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 1.080594326879784,
434
+ "grad_norm": 0.18488819323563785,
435
+ "learning_rate": 1.0330137426761136e-05,
436
+ "loss": 0.9832,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 1.0986042323277803,
441
+ "grad_norm": 0.18816750176239094,
442
+ "learning_rate": 1.0015723680475847e-05,
443
+ "loss": 0.9894,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 1.1166141377757768,
448
+ "grad_norm": 0.17955677324222288,
449
+ "learning_rate": 9.701294385737471e-06,
450
+ "loss": 0.9915,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 1.134624043223773,
455
+ "grad_norm": 0.1860265308958814,
456
+ "learning_rate": 9.38716046778684e-06,
457
+ "loss": 0.9702,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 1.1526339486717694,
462
+ "grad_norm": 0.18388832465015095,
463
+ "learning_rate": 9.073632559779731e-06,
464
+ "loss": 0.981,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 1.170643854119766,
469
+ "grad_norm": 0.1788808037903788,
470
+ "learning_rate": 8.76102069561545e-06,
471
+ "loss": 0.9812,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 1.1886537595677622,
476
+ "grad_norm": 0.178574258193573,
477
+ "learning_rate": 8.449634003358022e-06,
478
+ "loss": 0.9836,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 1.2066636650157587,
483
+ "grad_norm": 0.18306275208917874,
484
+ "learning_rate": 8.13978039955308e-06,
485
+ "loss": 0.9831,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 1.224673570463755,
490
+ "grad_norm": 0.18918449761657866,
491
+ "learning_rate": 7.831766284742807e-06,
492
+ "loss": 0.9846,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 1.2426834759117515,
497
+ "grad_norm": 0.1855294097858111,
498
+ "learning_rate": 7.525896240479977e-06,
499
+ "loss": 0.9744,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 1.2606933813597478,
504
+ "grad_norm": 0.20185722671737671,
505
+ "learning_rate": 7.222472728140695e-06,
506
+ "loss": 0.9859,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 1.2787032868077444,
511
+ "grad_norm": 0.192482680701874,
512
+ "learning_rate": 6.921795789833723e-06,
513
+ "loss": 0.9886,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 1.2967131922557407,
518
+ "grad_norm": 0.1795264410436132,
519
+ "learning_rate": 6.624162751702077e-06,
520
+ "loss": 0.9854,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 1.314723097703737,
525
+ "grad_norm": 0.18175034374465715,
526
+ "learning_rate": 6.329867929910347e-06,
527
+ "loss": 0.9836,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 1.3327330031517335,
532
+ "grad_norm": 0.18874834848612046,
533
+ "learning_rate": 6.039202339608432e-06,
534
+ "loss": 0.9792,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 1.3507429085997298,
539
+ "grad_norm": 0.1843992268180646,
540
+ "learning_rate": 5.752453407159521e-06,
541
+ "loss": 0.9794,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 1.3687528140477263,
546
+ "grad_norm": 0.17887298529702766,
547
+ "learning_rate": 5.469904685916861e-06,
548
+ "loss": 0.985,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 1.3867627194957226,
553
+ "grad_norm": 0.17428526152643262,
554
+ "learning_rate": 5.1918355758303515e-06,
555
+ "loss": 0.9793,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 1.4047726249437191,
560
+ "grad_norm": 0.17579200186681246,
561
+ "learning_rate": 4.918521047160309e-06,
562
+ "loss": 0.9797,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 1.4227825303917154,
567
+ "grad_norm": 0.17482239840906302,
568
+ "learning_rate": 4.650231368571486e-06,
569
+ "loss": 0.967,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 1.440792435839712,
574
+ "grad_norm": 0.1846232978343707,
575
+ "learning_rate": 4.387231839876349e-06,
576
+ "loss": 0.9795,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 1.4588023412877082,
581
+ "grad_norm": 0.17722992573590463,
582
+ "learning_rate": 4.1297825296918145e-06,
583
+ "loss": 0.9749,
584
+ "step": 810
585
+ },
586
+ {
587
+ "epoch": 1.4768122467357045,
588
+ "grad_norm": 0.18740341744136668,
589
+ "learning_rate": 3.878138018268867e-06,
590
+ "loss": 0.9846,
591
+ "step": 820
592
+ },
593
+ {
594
+ "epoch": 1.494822152183701,
595
+ "grad_norm": 0.18222853458564284,
596
+ "learning_rate": 3.6325471457493956e-06,
597
+ "loss": 0.9869,
598
+ "step": 830
599
+ },
600
+ {
601
+ "epoch": 1.5128320576316976,
602
+ "grad_norm": 0.17604705966690576,
603
+ "learning_rate": 3.3932527660991877e-06,
604
+ "loss": 0.9791,
605
+ "step": 840
606
+ },
607
+ {
608
+ "epoch": 1.530841963079694,
609
+ "grad_norm": 0.17129760493794519,
610
+ "learning_rate": 3.160491506960344e-06,
611
+ "loss": 0.9803,
612
+ "step": 850
613
+ },
614
+ {
615
+ "epoch": 1.5488518685276902,
616
+ "grad_norm": 0.17743331668980605,
617
+ "learning_rate": 2.934493535660677e-06,
618
+ "loss": 0.9849,
619
+ "step": 860
620
+ },
621
+ {
622
+ "epoch": 1.5668617739756865,
623
+ "grad_norm": 0.17206764149200166,
624
+ "learning_rate": 2.715482331611393e-06,
625
+ "loss": 0.9799,
626
+ "step": 870
627
+ },
628
+ {
629
+ "epoch": 1.584871679423683,
630
+ "grad_norm": 0.17401212500860314,
631
+ "learning_rate": 2.5036744653181755e-06,
632
+ "loss": 0.977,
633
+ "step": 880
634
+ },
635
+ {
636
+ "epoch": 1.6028815848716795,
637
+ "grad_norm": 0.1864847419394007,
638
+ "learning_rate": 2.29927938422419e-06,
639
+ "loss": 0.9824,
640
+ "step": 890
641
+ },
642
+ {
643
+ "epoch": 1.6208914903196758,
644
+ "grad_norm": 0.1794444567972596,
645
+ "learning_rate": 2.102499205596743e-06,
646
+ "loss": 0.9711,
647
+ "step": 900
648
+ },
649
+ {
650
+ "epoch": 1.6389013957676721,
651
+ "grad_norm": 0.17867218012908836,
652
+ "learning_rate": 1.913528516662452e-06,
653
+ "loss": 0.9904,
654
+ "step": 910
655
+ },
656
+ {
657
+ "epoch": 1.6569113012156687,
658
+ "grad_norm": 0.17632077595030837,
659
+ "learning_rate": 1.7325541821885383e-06,
660
+ "loss": 0.9846,
661
+ "step": 920
662
+ },
663
+ {
664
+ "epoch": 1.6749212066636652,
665
+ "grad_norm": 0.1749399044308501,
666
+ "learning_rate": 1.5597551597004968e-06,
667
+ "loss": 0.9759,
668
+ "step": 930
669
+ },
670
+ {
671
+ "epoch": 1.6929311121116615,
672
+ "grad_norm": 0.1765753424915534,
673
+ "learning_rate": 1.3953023225189243e-06,
674
+ "loss": 0.9947,
675
+ "step": 940
676
+ },
677
+ {
678
+ "epoch": 1.7109410175596578,
679
+ "grad_norm": 0.17249203862535814,
680
+ "learning_rate": 1.23935829079042e-06,
681
+ "loss": 0.987,
682
+ "step": 950
683
+ },
684
+ {
685
+ "epoch": 1.728950923007654,
686
+ "grad_norm": 0.17083864764712672,
687
+ "learning_rate": 1.0920772706797166e-06,
688
+ "loss": 0.9655,
689
+ "step": 960
690
+ },
691
+ {
692
+ "epoch": 1.7469608284556506,
693
+ "grad_norm": 0.17684409732386205,
694
+ "learning_rate": 9.536049018820193e-07,
695
+ "loss": 0.9741,
696
+ "step": 970
697
+ },
698
+ {
699
+ "epoch": 1.7649707339036471,
700
+ "grad_norm": 0.17089785327384419,
701
+ "learning_rate": 8.240781136063348e-07,
702
+ "loss": 0.9699,
703
+ "step": 980
704
+ },
705
+ {
706
+ "epoch": 1.7829806393516434,
707
+ "grad_norm": 0.17526364109022735,
708
+ "learning_rate": 7.03624989172228e-07,
709
+ "loss": 0.9884,
710
+ "step": 990
711
+ },
712
+ {
713
+ "epoch": 1.8009905447996397,
714
+ "grad_norm": 0.17477084621817432,
715
+ "learning_rate": 5.923646393538907e-07,
716
+ "loss": 0.9762,
717
+ "step": 1000
718
+ },
719
+ {
720
+ "epoch": 1.8190004502476362,
721
+ "grad_norm": 0.17586999782551316,
722
+ "learning_rate": 4.904070845967467e-07,
723
+ "loss": 0.9763,
724
+ "step": 1010
725
+ },
726
+ {
727
+ "epoch": 1.8370103556956328,
728
+ "grad_norm": 0.17351730726317385,
729
+ "learning_rate": 3.97853146223105e-07,
730
+ "loss": 0.9759,
731
+ "step": 1020
732
+ },
733
+ {
734
+ "epoch": 1.855020261143629,
735
+ "grad_norm": 0.17245529759955128,
736
+ "learning_rate": 3.1479434673440167e-07,
737
+ "loss": 0.9762,
738
+ "step": 1030
739
+ },
740
+ {
741
+ "epoch": 1.8730301665916254,
742
+ "grad_norm": 0.17074305784570018,
743
+ "learning_rate": 2.4131281930864006e-07,
744
+ "loss": 0.9778,
745
+ "step": 1040
746
+ },
747
+ {
748
+ "epoch": 1.8910400720396217,
749
+ "grad_norm": 0.1699923858478813,
750
+ "learning_rate": 1.7748122658251877e-07,
751
+ "loss": 0.9898,
752
+ "step": 1050
753
+ },
754
+ {
755
+ "epoch": 1.9090499774876182,
756
+ "grad_norm": 0.17560912690299982,
757
+ "learning_rate": 1.2336268879856728e-07,
758
+ "loss": 0.9764,
759
+ "step": 1060
760
+ },
761
+ {
762
+ "epoch": 1.9270598829356147,
763
+ "grad_norm": 0.17172727854066605,
764
+ "learning_rate": 7.901072138831512e-08,
765
+ "loss": 0.9908,
766
+ "step": 1070
767
+ },
768
+ {
769
+ "epoch": 1.945069788383611,
770
+ "grad_norm": 0.17416282443666103,
771
+ "learning_rate": 4.44691820532539e-08,
772
+ "loss": 0.9765,
773
+ "step": 1080
774
+ },
775
+ {
776
+ "epoch": 1.9630796938316073,
777
+ "grad_norm": 0.1719348511857924,
778
+ "learning_rate": 1.977222739588891e-08,
779
+ "loss": 0.9784,
780
+ "step": 1090
781
+ },
782
+ {
783
+ "epoch": 1.9810895992796038,
784
+ "grad_norm": 0.1744398791007269,
785
+ "learning_rate": 4.944279143784814e-09,
786
+ "loss": 0.9841,
787
+ "step": 1100
788
+ },
789
+ {
790
+ "epoch": 1.9990995047276003,
791
+ "grad_norm": 0.17567978804671192,
792
+ "learning_rate": 0.0,
793
+ "loss": 0.9698,
794
+ "step": 1110
795
+ },
796
+ {
797
+ "epoch": 1.9990995047276003,
798
+ "eval_loss": 1.0319925546646118,
799
+ "eval_runtime": 659.9931,
800
+ "eval_samples_per_second": 11.917,
801
+ "eval_steps_per_second": 0.745,
802
+ "step": 1110
803
+ },
804
+ {
805
+ "epoch": 1.9990995047276003,
806
+ "step": 1110,
807
+ "total_flos": 929619498762240.0,
808
+ "train_loss": 1.0262996888375497,
809
+ "train_runtime": 54210.2289,
810
+ "train_samples_per_second": 2.622,
811
+ "train_steps_per_second": 0.02
812
+ }
813
+ ],
814
+ "logging_steps": 10,
815
+ "max_steps": 1110,
816
+ "num_input_tokens_seen": 0,
817
+ "num_train_epochs": 2,
818
+ "save_steps": 100,
819
+ "stateful_callbacks": {
820
+ "TrainerControl": {
821
+ "args": {
822
+ "should_epoch_stop": false,
823
+ "should_evaluate": false,
824
+ "should_log": false,
825
+ "should_save": true,
826
+ "should_training_stop": true
827
+ },
828
+ "attributes": {}
829
+ }
830
+ },
831
+ "total_flos": 929619498762240.0,
832
+ "train_batch_size": 1,
833
+ "trial_name": null,
834
+ "trial_params": null
835
+ }