chansung commited on
Commit
c93032c
1 Parent(s): 370cfef

Model save

Browse files
Files changed (4) hide show
  1. README.md +69 -0
  2. all_results.json +9 -0
  3. train_results.json +9 -0
  4. trainer_state.json +953 -0
README.md ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: mistralai/Mistral-7B-v0.3
3
+ datasets:
4
+ - generator
5
+ library_name: peft
6
+ license: apache-2.0
7
+ tags:
8
+ - trl
9
+ - sft
10
+ - generated_from_trainer
11
+ model-index:
12
+ - name: mistral-7b-0.3-gpt4o_100k_coding-lora
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_coding-lora
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: 1.0356
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
+ | 0.4351 | 0.9992 | 640 | 1.0356 |
61
+
62
+
63
+ ### Framework versions
64
+
65
+ - PEFT 0.13.0
66
+ - Transformers 4.45.1
67
+ - Pytorch 2.4.1+cu121
68
+ - Datasets 3.0.1
69
+ - Tokenizers 0.20.0
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 0.9992193598750976,
3
+ "total_flos": 8.955537329204756e+17,
4
+ "train_loss": 0.4811625644564629,
5
+ "train_runtime": 4514.48,
6
+ "train_samples": 116368,
7
+ "train_samples_per_second": 4.539,
8
+ "train_steps_per_second": 0.142
9
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 0.9992193598750976,
3
+ "total_flos": 8.955537329204756e+17,
4
+ "train_loss": 0.4811625644564629,
5
+ "train_runtime": 4514.48,
6
+ "train_samples": 116368,
7
+ "train_samples_per_second": 4.539,
8
+ "train_steps_per_second": 0.142
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,953 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.9992193598750976,
5
+ "eval_steps": 500,
6
+ "global_step": 640,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.00156128024980484,
13
+ "grad_norm": 4.679694652557373,
14
+ "learning_rate": 3.125e-06,
15
+ "loss": 1.3,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.0078064012490242,
20
+ "grad_norm": 4.700327396392822,
21
+ "learning_rate": 1.5625e-05,
22
+ "loss": 1.2534,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.0156128024980484,
27
+ "grad_norm": 4.184088230133057,
28
+ "learning_rate": 3.125e-05,
29
+ "loss": 1.1414,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 0.0234192037470726,
34
+ "grad_norm": 3.3843367099761963,
35
+ "learning_rate": 4.6875e-05,
36
+ "loss": 0.9791,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.0312256049960968,
41
+ "grad_norm": 1.38542640209198,
42
+ "learning_rate": 6.25e-05,
43
+ "loss": 0.8613,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 0.039032006245121,
48
+ "grad_norm": 0.8451548218727112,
49
+ "learning_rate": 7.8125e-05,
50
+ "loss": 0.7475,
51
+ "step": 25
52
+ },
53
+ {
54
+ "epoch": 0.0468384074941452,
55
+ "grad_norm": 0.6994028091430664,
56
+ "learning_rate": 9.375e-05,
57
+ "loss": 0.6857,
58
+ "step": 30
59
+ },
60
+ {
61
+ "epoch": 0.0546448087431694,
62
+ "grad_norm": 0.7693173885345459,
63
+ "learning_rate": 0.000109375,
64
+ "loss": 0.632,
65
+ "step": 35
66
+ },
67
+ {
68
+ "epoch": 0.0624512099921936,
69
+ "grad_norm": 1.0630561113357544,
70
+ "learning_rate": 0.000125,
71
+ "loss": 0.6101,
72
+ "step": 40
73
+ },
74
+ {
75
+ "epoch": 0.0702576112412178,
76
+ "grad_norm": 0.9208365678787231,
77
+ "learning_rate": 0.00014062500000000002,
78
+ "loss": 0.5723,
79
+ "step": 45
80
+ },
81
+ {
82
+ "epoch": 0.078064012490242,
83
+ "grad_norm": 0.8858307003974915,
84
+ "learning_rate": 0.00015625,
85
+ "loss": 0.5483,
86
+ "step": 50
87
+ },
88
+ {
89
+ "epoch": 0.0858704137392662,
90
+ "grad_norm": 0.5890257358551025,
91
+ "learning_rate": 0.00017187500000000002,
92
+ "loss": 0.5341,
93
+ "step": 55
94
+ },
95
+ {
96
+ "epoch": 0.0936768149882904,
97
+ "grad_norm": 0.8059648275375366,
98
+ "learning_rate": 0.0001875,
99
+ "loss": 0.5393,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 0.1014832162373146,
104
+ "grad_norm": 0.6733461618423462,
105
+ "learning_rate": 0.00019999851261394218,
106
+ "loss": 0.5324,
107
+ "step": 65
108
+ },
109
+ {
110
+ "epoch": 0.1092896174863388,
111
+ "grad_norm": 0.5950048565864563,
112
+ "learning_rate": 0.00019994645874763658,
113
+ "loss": 0.5186,
114
+ "step": 70
115
+ },
116
+ {
117
+ "epoch": 0.117096018735363,
118
+ "grad_norm": 0.7100915908813477,
119
+ "learning_rate": 0.00019982007981886847,
120
+ "loss": 0.5022,
121
+ "step": 75
122
+ },
123
+ {
124
+ "epoch": 0.1249024199843872,
125
+ "grad_norm": 0.38926348090171814,
126
+ "learning_rate": 0.00019961946980917456,
127
+ "loss": 0.5032,
128
+ "step": 80
129
+ },
130
+ {
131
+ "epoch": 0.1327088212334114,
132
+ "grad_norm": 0.32585254311561584,
133
+ "learning_rate": 0.00019934477790194445,
134
+ "loss": 0.4926,
135
+ "step": 85
136
+ },
137
+ {
138
+ "epoch": 0.1405152224824356,
139
+ "grad_norm": 0.37917089462280273,
140
+ "learning_rate": 0.00019899620837148077,
141
+ "loss": 0.4963,
142
+ "step": 90
143
+ },
144
+ {
145
+ "epoch": 0.1483216237314598,
146
+ "grad_norm": 0.4928993582725525,
147
+ "learning_rate": 0.0001985740204310909,
148
+ "loss": 0.4934,
149
+ "step": 95
150
+ },
151
+ {
152
+ "epoch": 0.156128024980484,
153
+ "grad_norm": 0.41284865140914917,
154
+ "learning_rate": 0.00019807852804032305,
155
+ "loss": 0.4988,
156
+ "step": 100
157
+ },
158
+ {
159
+ "epoch": 0.16393442622950818,
160
+ "grad_norm": 0.3852866291999817,
161
+ "learning_rate": 0.00019751009967149087,
162
+ "loss": 0.4861,
163
+ "step": 105
164
+ },
165
+ {
166
+ "epoch": 0.1717408274785324,
167
+ "grad_norm": 0.34940582513809204,
168
+ "learning_rate": 0.00019686915803565934,
169
+ "loss": 0.4846,
170
+ "step": 110
171
+ },
172
+ {
173
+ "epoch": 0.1795472287275566,
174
+ "grad_norm": 0.379056453704834,
175
+ "learning_rate": 0.0001961561797682962,
176
+ "loss": 0.4716,
177
+ "step": 115
178
+ },
179
+ {
180
+ "epoch": 0.1873536299765808,
181
+ "grad_norm": 0.378898024559021,
182
+ "learning_rate": 0.0001953716950748227,
183
+ "loss": 0.4862,
184
+ "step": 120
185
+ },
186
+ {
187
+ "epoch": 0.195160031225605,
188
+ "grad_norm": 0.5000614523887634,
189
+ "learning_rate": 0.0001945162873363268,
190
+ "loss": 0.4758,
191
+ "step": 125
192
+ },
193
+ {
194
+ "epoch": 0.2029664324746292,
195
+ "grad_norm": 0.43147751688957214,
196
+ "learning_rate": 0.0001935905926757326,
197
+ "loss": 0.477,
198
+ "step": 130
199
+ },
200
+ {
201
+ "epoch": 0.2107728337236534,
202
+ "grad_norm": 0.41404739022254944,
203
+ "learning_rate": 0.00019259529948474833,
204
+ "loss": 0.4773,
205
+ "step": 135
206
+ },
207
+ {
208
+ "epoch": 0.2185792349726776,
209
+ "grad_norm": 0.4057549834251404,
210
+ "learning_rate": 0.00019153114791194473,
211
+ "loss": 0.4777,
212
+ "step": 140
213
+ },
214
+ {
215
+ "epoch": 0.2263856362217018,
216
+ "grad_norm": 0.3698827624320984,
217
+ "learning_rate": 0.00019039892931234435,
218
+ "loss": 0.4911,
219
+ "step": 145
220
+ },
221
+ {
222
+ "epoch": 0.234192037470726,
223
+ "grad_norm": 0.4173194169998169,
224
+ "learning_rate": 0.00018919948565893142,
225
+ "loss": 0.4722,
226
+ "step": 150
227
+ },
228
+ {
229
+ "epoch": 0.2419984387197502,
230
+ "grad_norm": 0.5116965174674988,
231
+ "learning_rate": 0.00018793370891651972,
232
+ "loss": 0.4766,
233
+ "step": 155
234
+ },
235
+ {
236
+ "epoch": 0.2498048399687744,
237
+ "grad_norm": 0.33461353182792664,
238
+ "learning_rate": 0.00018660254037844388,
239
+ "loss": 0.4833,
240
+ "step": 160
241
+ },
242
+ {
243
+ "epoch": 0.2576112412177986,
244
+ "grad_norm": 0.45119625329971313,
245
+ "learning_rate": 0.00018520696996656788,
246
+ "loss": 0.4633,
247
+ "step": 165
248
+ },
249
+ {
250
+ "epoch": 0.2654176424668228,
251
+ "grad_norm": 0.5022788047790527,
252
+ "learning_rate": 0.0001837480354951308,
253
+ "loss": 0.4539,
254
+ "step": 170
255
+ },
256
+ {
257
+ "epoch": 0.273224043715847,
258
+ "grad_norm": 0.37766632437705994,
259
+ "learning_rate": 0.00018222682189897752,
260
+ "loss": 0.4615,
261
+ "step": 175
262
+ },
263
+ {
264
+ "epoch": 0.2810304449648712,
265
+ "grad_norm": 0.34759408235549927,
266
+ "learning_rate": 0.00018064446042674828,
267
+ "loss": 0.4621,
268
+ "step": 180
269
+ },
270
+ {
271
+ "epoch": 0.2888368462138954,
272
+ "grad_norm": 0.5189246535301208,
273
+ "learning_rate": 0.0001790021277996269,
274
+ "loss": 0.4644,
275
+ "step": 185
276
+ },
277
+ {
278
+ "epoch": 0.2966432474629196,
279
+ "grad_norm": 0.4171973466873169,
280
+ "learning_rate": 0.0001773010453362737,
281
+ "loss": 0.4733,
282
+ "step": 190
283
+ },
284
+ {
285
+ "epoch": 0.3044496487119438,
286
+ "grad_norm": 0.3633343577384949,
287
+ "learning_rate": 0.00017554247804459316,
288
+ "loss": 0.4647,
289
+ "step": 195
290
+ },
291
+ {
292
+ "epoch": 0.312256049960968,
293
+ "grad_norm": 0.3888181746006012,
294
+ "learning_rate": 0.0001737277336810124,
295
+ "loss": 0.4666,
296
+ "step": 200
297
+ },
298
+ {
299
+ "epoch": 0.3200624512099922,
300
+ "grad_norm": 0.35728883743286133,
301
+ "learning_rate": 0.0001718581617779698,
302
+ "loss": 0.4612,
303
+ "step": 205
304
+ },
305
+ {
306
+ "epoch": 0.32786885245901637,
307
+ "grad_norm": 0.3724042475223541,
308
+ "learning_rate": 0.00016993515264033672,
309
+ "loss": 0.4718,
310
+ "step": 210
311
+ },
312
+ {
313
+ "epoch": 0.3356752537080406,
314
+ "grad_norm": 0.3650490641593933,
315
+ "learning_rate": 0.00016796013631151897,
316
+ "loss": 0.4519,
317
+ "step": 215
318
+ },
319
+ {
320
+ "epoch": 0.3434816549570648,
321
+ "grad_norm": 0.3484860062599182,
322
+ "learning_rate": 0.00016593458151000688,
323
+ "loss": 0.4506,
324
+ "step": 220
325
+ },
326
+ {
327
+ "epoch": 0.351288056206089,
328
+ "grad_norm": 0.32455822825431824,
329
+ "learning_rate": 0.00016385999453716454,
330
+ "loss": 0.4456,
331
+ "step": 225
332
+ },
333
+ {
334
+ "epoch": 0.3590944574551132,
335
+ "grad_norm": 0.4763931334018707,
336
+ "learning_rate": 0.00016173791815707051,
337
+ "loss": 0.4584,
338
+ "step": 230
339
+ },
340
+ {
341
+ "epoch": 0.3669008587041374,
342
+ "grad_norm": 0.37389329075813293,
343
+ "learning_rate": 0.00015956993044924334,
344
+ "loss": 0.4502,
345
+ "step": 235
346
+ },
347
+ {
348
+ "epoch": 0.3747072599531616,
349
+ "grad_norm": 0.5365679860115051,
350
+ "learning_rate": 0.0001573576436351046,
351
+ "loss": 0.462,
352
+ "step": 240
353
+ },
354
+ {
355
+ "epoch": 0.3825136612021858,
356
+ "grad_norm": 0.3317677080631256,
357
+ "learning_rate": 0.0001551027028790524,
358
+ "loss": 0.4494,
359
+ "step": 245
360
+ },
361
+ {
362
+ "epoch": 0.39032006245121,
363
+ "grad_norm": 0.3652748763561249,
364
+ "learning_rate": 0.0001528067850650368,
365
+ "loss": 0.4608,
366
+ "step": 250
367
+ },
368
+ {
369
+ "epoch": 0.3981264637002342,
370
+ "grad_norm": 0.3600890636444092,
371
+ "learning_rate": 0.0001504715975495472,
372
+ "loss": 0.4535,
373
+ "step": 255
374
+ },
375
+ {
376
+ "epoch": 0.4059328649492584,
377
+ "grad_norm": 0.3843066990375519,
378
+ "learning_rate": 0.00014809887689193877,
379
+ "loss": 0.4471,
380
+ "step": 260
381
+ },
382
+ {
383
+ "epoch": 0.4137392661982826,
384
+ "grad_norm": 0.42747819423675537,
385
+ "learning_rate": 0.00014569038756304207,
386
+ "loss": 0.4667,
387
+ "step": 265
388
+ },
389
+ {
390
+ "epoch": 0.4215456674473068,
391
+ "grad_norm": 0.354117214679718,
392
+ "learning_rate": 0.00014324792063301662,
393
+ "loss": 0.4481,
394
+ "step": 270
395
+ },
396
+ {
397
+ "epoch": 0.42935206869633097,
398
+ "grad_norm": 0.3524267375469208,
399
+ "learning_rate": 0.00014077329243942369,
400
+ "loss": 0.4511,
401
+ "step": 275
402
+ },
403
+ {
404
+ "epoch": 0.4371584699453552,
405
+ "grad_norm": 0.41965606808662415,
406
+ "learning_rate": 0.000138268343236509,
407
+ "loss": 0.4519,
408
+ "step": 280
409
+ },
410
+ {
411
+ "epoch": 0.4449648711943794,
412
+ "grad_norm": 0.338069349527359,
413
+ "learning_rate": 0.00013573493582670003,
414
+ "loss": 0.4448,
415
+ "step": 285
416
+ },
417
+ {
418
+ "epoch": 0.4527712724434036,
419
+ "grad_norm": 0.44666725397109985,
420
+ "learning_rate": 0.00013317495417533524,
421
+ "loss": 0.451,
422
+ "step": 290
423
+ },
424
+ {
425
+ "epoch": 0.4605776736924278,
426
+ "grad_norm": 0.3880995213985443,
427
+ "learning_rate": 0.00013059030200965536,
428
+ "loss": 0.4402,
429
+ "step": 295
430
+ },
431
+ {
432
+ "epoch": 0.468384074941452,
433
+ "grad_norm": 0.39150258898735046,
434
+ "learning_rate": 0.00012798290140309923,
435
+ "loss": 0.4337,
436
+ "step": 300
437
+ },
438
+ {
439
+ "epoch": 0.47619047619047616,
440
+ "grad_norm": 0.3439214825630188,
441
+ "learning_rate": 0.00012535469134595595,
442
+ "loss": 0.4487,
443
+ "step": 305
444
+ },
445
+ {
446
+ "epoch": 0.4839968774395004,
447
+ "grad_norm": 0.44355058670043945,
448
+ "learning_rate": 0.00012270762630343734,
449
+ "loss": 0.4448,
450
+ "step": 310
451
+ },
452
+ {
453
+ "epoch": 0.4918032786885246,
454
+ "grad_norm": 0.36502349376678467,
455
+ "learning_rate": 0.00012004367476224206,
456
+ "loss": 0.4488,
457
+ "step": 315
458
+ },
459
+ {
460
+ "epoch": 0.4996096799375488,
461
+ "grad_norm": 0.4983600080013275,
462
+ "learning_rate": 0.00011736481776669306,
463
+ "loss": 0.4308,
464
+ "step": 320
465
+ },
466
+ {
467
+ "epoch": 0.507416081186573,
468
+ "grad_norm": 0.5701916217803955,
469
+ "learning_rate": 0.00011467304744553618,
470
+ "loss": 0.4425,
471
+ "step": 325
472
+ },
473
+ {
474
+ "epoch": 0.5152224824355972,
475
+ "grad_norm": 0.35758593678474426,
476
+ "learning_rate": 0.00011197036553049625,
477
+ "loss": 0.4425,
478
+ "step": 330
479
+ },
480
+ {
481
+ "epoch": 0.5230288836846214,
482
+ "grad_norm": 0.3369368612766266,
483
+ "learning_rate": 0.00010925878186769158,
484
+ "loss": 0.4444,
485
+ "step": 335
486
+ },
487
+ {
488
+ "epoch": 0.5308352849336456,
489
+ "grad_norm": 0.35938557982444763,
490
+ "learning_rate": 0.00010654031292301432,
491
+ "loss": 0.4363,
492
+ "step": 340
493
+ },
494
+ {
495
+ "epoch": 0.5386416861826698,
496
+ "grad_norm": 0.34216657280921936,
497
+ "learning_rate": 0.00010381698028258817,
498
+ "loss": 0.4421,
499
+ "step": 345
500
+ },
501
+ {
502
+ "epoch": 0.546448087431694,
503
+ "grad_norm": 0.36699554324150085,
504
+ "learning_rate": 0.00010109080914941824,
505
+ "loss": 0.4558,
506
+ "step": 350
507
+ },
508
+ {
509
+ "epoch": 0.5542544886807181,
510
+ "grad_norm": 0.36921554803848267,
511
+ "learning_rate": 9.836382683735132e-05,
512
+ "loss": 0.4514,
513
+ "step": 355
514
+ },
515
+ {
516
+ "epoch": 0.5620608899297423,
517
+ "grad_norm": 0.3594672977924347,
518
+ "learning_rate": 9.563806126346642e-05,
519
+ "loss": 0.4518,
520
+ "step": 360
521
+ },
522
+ {
523
+ "epoch": 0.5698672911787666,
524
+ "grad_norm": 0.37948325276374817,
525
+ "learning_rate": 9.29155394400166e-05,
526
+ "loss": 0.45,
527
+ "step": 365
528
+ },
529
+ {
530
+ "epoch": 0.5776736924277908,
531
+ "grad_norm": 0.3441675007343292,
532
+ "learning_rate": 9.019828596704394e-05,
533
+ "loss": 0.4425,
534
+ "step": 370
535
+ },
536
+ {
537
+ "epoch": 0.585480093676815,
538
+ "grad_norm": 0.3524007499217987,
539
+ "learning_rate": 8.74883215267881e-05,
540
+ "loss": 0.434,
541
+ "step": 375
542
+ },
543
+ {
544
+ "epoch": 0.5932864949258392,
545
+ "grad_norm": 0.38553693890571594,
546
+ "learning_rate": 8.478766138100834e-05,
547
+ "loss": 0.4352,
548
+ "step": 380
549
+ },
550
+ {
551
+ "epoch": 0.6010928961748634,
552
+ "grad_norm": 0.3395775854587555,
553
+ "learning_rate": 8.209831387233676e-05,
554
+ "loss": 0.4472,
555
+ "step": 385
556
+ },
557
+ {
558
+ "epoch": 0.6088992974238876,
559
+ "grad_norm": 0.354542076587677,
560
+ "learning_rate": 7.942227893077652e-05,
561
+ "loss": 0.4418,
562
+ "step": 390
563
+ },
564
+ {
565
+ "epoch": 0.6167056986729118,
566
+ "grad_norm": 0.3631182610988617,
567
+ "learning_rate": 7.676154658645656e-05,
568
+ "loss": 0.4361,
569
+ "step": 395
570
+ },
571
+ {
572
+ "epoch": 0.624512099921936,
573
+ "grad_norm": 0.35347339510917664,
574
+ "learning_rate": 7.411809548974792e-05,
575
+ "loss": 0.4348,
576
+ "step": 400
577
+ },
578
+ {
579
+ "epoch": 0.6323185011709602,
580
+ "grad_norm": 0.4688993990421295,
581
+ "learning_rate": 7.149389143984295e-05,
582
+ "loss": 0.4464,
583
+ "step": 405
584
+ },
585
+ {
586
+ "epoch": 0.6401249024199844,
587
+ "grad_norm": 0.3508263826370239,
588
+ "learning_rate": 6.889088592289093e-05,
589
+ "loss": 0.4428,
590
+ "step": 410
591
+ },
592
+ {
593
+ "epoch": 0.6479313036690086,
594
+ "grad_norm": 0.3598111867904663,
595
+ "learning_rate": 6.6311014660778e-05,
596
+ "loss": 0.4442,
597
+ "step": 415
598
+ },
599
+ {
600
+ "epoch": 0.6557377049180327,
601
+ "grad_norm": 0.35423406958580017,
602
+ "learning_rate": 6.375619617162985e-05,
603
+ "loss": 0.4442,
604
+ "step": 420
605
+ },
606
+ {
607
+ "epoch": 0.663544106167057,
608
+ "grad_norm": 0.32423022389411926,
609
+ "learning_rate": 6.122833034310793e-05,
610
+ "loss": 0.4385,
611
+ "step": 425
612
+ },
613
+ {
614
+ "epoch": 0.6713505074160812,
615
+ "grad_norm": 0.34300023317337036,
616
+ "learning_rate": 5.872929701956054e-05,
617
+ "loss": 0.4314,
618
+ "step": 430
619
+ },
620
+ {
621
+ "epoch": 0.6791569086651054,
622
+ "grad_norm": 0.355337530374527,
623
+ "learning_rate": 5.6260954604078585e-05,
624
+ "loss": 0.4437,
625
+ "step": 435
626
+ },
627
+ {
628
+ "epoch": 0.6869633099141296,
629
+ "grad_norm": 0.3726743161678314,
630
+ "learning_rate": 5.382513867649663e-05,
631
+ "loss": 0.4324,
632
+ "step": 440
633
+ },
634
+ {
635
+ "epoch": 0.6947697111631538,
636
+ "grad_norm": 0.36385682225227356,
637
+ "learning_rate": 5.142366062836599e-05,
638
+ "loss": 0.4398,
639
+ "step": 445
640
+ },
641
+ {
642
+ "epoch": 0.702576112412178,
643
+ "grad_norm": 0.34697508811950684,
644
+ "learning_rate": 4.9058306315915826e-05,
645
+ "loss": 0.4496,
646
+ "step": 450
647
+ },
648
+ {
649
+ "epoch": 0.7103825136612022,
650
+ "grad_norm": 0.37256768345832825,
651
+ "learning_rate": 4.6730834732003104e-05,
652
+ "loss": 0.4432,
653
+ "step": 455
654
+ },
655
+ {
656
+ "epoch": 0.7181889149102264,
657
+ "grad_norm": 0.36101406812667847,
658
+ "learning_rate": 4.444297669803981e-05,
659
+ "loss": 0.4506,
660
+ "step": 460
661
+ },
662
+ {
663
+ "epoch": 0.7259953161592506,
664
+ "grad_norm": 0.3514850437641144,
665
+ "learning_rate": 4.219643357686967e-05,
666
+ "loss": 0.4321,
667
+ "step": 465
668
+ },
669
+ {
670
+ "epoch": 0.7338017174082748,
671
+ "grad_norm": 0.36538928747177124,
672
+ "learning_rate": 3.999287600755192e-05,
673
+ "loss": 0.4278,
674
+ "step": 470
675
+ },
676
+ {
677
+ "epoch": 0.741608118657299,
678
+ "grad_norm": 0.3825438320636749,
679
+ "learning_rate": 3.783394266299228e-05,
680
+ "loss": 0.4368,
681
+ "step": 475
682
+ },
683
+ {
684
+ "epoch": 0.7494145199063232,
685
+ "grad_norm": 0.4204176068305969,
686
+ "learning_rate": 3.5721239031346066e-05,
687
+ "loss": 0.443,
688
+ "step": 480
689
+ },
690
+ {
691
+ "epoch": 0.7572209211553473,
692
+ "grad_norm": 0.3882097005844116,
693
+ "learning_rate": 3.365633622209891e-05,
694
+ "loss": 0.437,
695
+ "step": 485
696
+ },
697
+ {
698
+ "epoch": 0.7650273224043715,
699
+ "grad_norm": 0.38486871123313904,
700
+ "learning_rate": 3.164076979771287e-05,
701
+ "loss": 0.4362,
702
+ "step": 490
703
+ },
704
+ {
705
+ "epoch": 0.7728337236533958,
706
+ "grad_norm": 0.3774687647819519,
707
+ "learning_rate": 2.9676038631707593e-05,
708
+ "loss": 0.4341,
709
+ "step": 495
710
+ },
711
+ {
712
+ "epoch": 0.78064012490242,
713
+ "grad_norm": 0.3509750962257385,
714
+ "learning_rate": 2.776360379402445e-05,
715
+ "loss": 0.437,
716
+ "step": 500
717
+ },
718
+ {
719
+ "epoch": 0.7884465261514442,
720
+ "grad_norm": 0.4982232451438904,
721
+ "learning_rate": 2.5904887464504114e-05,
722
+ "loss": 0.4352,
723
+ "step": 505
724
+ },
725
+ {
726
+ "epoch": 0.7962529274004684,
727
+ "grad_norm": 0.3576488196849823,
728
+ "learning_rate": 2.4101271875283817e-05,
729
+ "loss": 0.4415,
730
+ "step": 510
731
+ },
732
+ {
733
+ "epoch": 0.8040593286494926,
734
+ "grad_norm": 0.3655681312084198,
735
+ "learning_rate": 2.2354098282902446e-05,
736
+ "loss": 0.4302,
737
+ "step": 515
738
+ },
739
+ {
740
+ "epoch": 0.8118657298985168,
741
+ "grad_norm": 0.34478428959846497,
742
+ "learning_rate": 2.0664665970876496e-05,
743
+ "loss": 0.4472,
744
+ "step": 520
745
+ },
746
+ {
747
+ "epoch": 0.819672131147541,
748
+ "grad_norm": 0.34541845321655273,
749
+ "learning_rate": 1.903423128348959e-05,
750
+ "loss": 0.4425,
751
+ "step": 525
752
+ },
753
+ {
754
+ "epoch": 0.8274785323965652,
755
+ "grad_norm": 0.3647783398628235,
756
+ "learning_rate": 1.7464006691513623e-05,
757
+ "loss": 0.4354,
758
+ "step": 530
759
+ },
760
+ {
761
+ "epoch": 0.8352849336455894,
762
+ "grad_norm": 0.3986523747444153,
763
+ "learning_rate": 1.595515989055618e-05,
764
+ "loss": 0.4483,
765
+ "step": 535
766
+ },
767
+ {
768
+ "epoch": 0.8430913348946136,
769
+ "grad_norm": 0.3613293766975403,
770
+ "learning_rate": 1.4508812932705363e-05,
771
+ "loss": 0.4412,
772
+ "step": 540
773
+ },
774
+ {
775
+ "epoch": 0.8508977361436377,
776
+ "grad_norm": 0.3343644142150879,
777
+ "learning_rate": 1.3126041392116772e-05,
778
+ "loss": 0.4445,
779
+ "step": 545
780
+ },
781
+ {
782
+ "epoch": 0.8587041373926619,
783
+ "grad_norm": 0.34114307165145874,
784
+ "learning_rate": 1.1807873565164506e-05,
785
+ "loss": 0.4483,
786
+ "step": 550
787
+ },
788
+ {
789
+ "epoch": 0.8665105386416861,
790
+ "grad_norm": 0.36701440811157227,
791
+ "learning_rate": 1.0555289705749483e-05,
792
+ "loss": 0.4351,
793
+ "step": 555
794
+ },
795
+ {
796
+ "epoch": 0.8743169398907104,
797
+ "grad_norm": 0.3947718143463135,
798
+ "learning_rate": 9.369221296335006e-06,
799
+ "loss": 0.4295,
800
+ "step": 560
801
+ },
802
+ {
803
+ "epoch": 0.8821233411397346,
804
+ "grad_norm": 0.3417580723762512,
805
+ "learning_rate": 8.250550355250875e-06,
806
+ "loss": 0.4296,
807
+ "step": 565
808
+ },
809
+ {
810
+ "epoch": 0.8899297423887588,
811
+ "grad_norm": 0.35511329770088196,
812
+ "learning_rate": 7.200108780781556e-06,
813
+ "loss": 0.45,
814
+ "step": 570
815
+ },
816
+ {
817
+ "epoch": 0.897736143637783,
818
+ "grad_norm": 0.3438127636909485,
819
+ "learning_rate": 6.218677732526035e-06,
820
+ "loss": 0.434,
821
+ "step": 575
822
+ },
823
+ {
824
+ "epoch": 0.9055425448868072,
825
+ "grad_norm": 0.35285842418670654,
826
+ "learning_rate": 5.306987050489442e-06,
827
+ "loss": 0.4339,
828
+ "step": 580
829
+ },
830
+ {
831
+ "epoch": 0.9133489461358314,
832
+ "grad_norm": 0.37091735005378723,
833
+ "learning_rate": 4.465714712338398e-06,
834
+ "loss": 0.4422,
835
+ "step": 585
836
+ },
837
+ {
838
+ "epoch": 0.9211553473848556,
839
+ "grad_norm": 0.3490866422653198,
840
+ "learning_rate": 3.6954863292237297e-06,
841
+ "loss": 0.437,
842
+ "step": 590
843
+ },
844
+ {
845
+ "epoch": 0.9289617486338798,
846
+ "grad_norm": 0.3426539897918701,
847
+ "learning_rate": 2.996874680545603e-06,
848
+ "loss": 0.4461,
849
+ "step": 595
850
+ },
851
+ {
852
+ "epoch": 0.936768149882904,
853
+ "grad_norm": 0.359111487865448,
854
+ "learning_rate": 2.3703992880066638e-06,
855
+ "loss": 0.4405,
856
+ "step": 600
857
+ },
858
+ {
859
+ "epoch": 0.9445745511319282,
860
+ "grad_norm": 0.35233959555625916,
861
+ "learning_rate": 1.8165260292704711e-06,
862
+ "loss": 0.4312,
863
+ "step": 605
864
+ },
865
+ {
866
+ "epoch": 0.9523809523809523,
867
+ "grad_norm": 0.3432738184928894,
868
+ "learning_rate": 1.3356667915121025e-06,
869
+ "loss": 0.4395,
870
+ "step": 610
871
+ },
872
+ {
873
+ "epoch": 0.9601873536299765,
874
+ "grad_norm": 0.3486664295196533,
875
+ "learning_rate": 9.281791651187366e-07,
876
+ "loss": 0.4326,
877
+ "step": 615
878
+ },
879
+ {
880
+ "epoch": 0.9679937548790007,
881
+ "grad_norm": 0.35764941573143005,
882
+ "learning_rate": 5.943661777680354e-07,
883
+ "loss": 0.4345,
884
+ "step": 620
885
+ },
886
+ {
887
+ "epoch": 0.975800156128025,
888
+ "grad_norm": 0.3644266128540039,
889
+ "learning_rate": 3.3447606908196817e-07,
890
+ "loss": 0.4317,
891
+ "step": 625
892
+ },
893
+ {
894
+ "epoch": 0.9836065573770492,
895
+ "grad_norm": 0.34483930468559265,
896
+ "learning_rate": 1.487021060236904e-07,
897
+ "loss": 0.4345,
898
+ "step": 630
899
+ },
900
+ {
901
+ "epoch": 0.9914129586260734,
902
+ "grad_norm": 0.36307430267333984,
903
+ "learning_rate": 3.7182439174832106e-08,
904
+ "loss": 0.4348,
905
+ "step": 635
906
+ },
907
+ {
908
+ "epoch": 0.9992193598750976,
909
+ "grad_norm": 0.3593035042285919,
910
+ "learning_rate": 0.0,
911
+ "loss": 0.4351,
912
+ "step": 640
913
+ },
914
+ {
915
+ "epoch": 0.9992193598750976,
916
+ "eval_loss": 1.0355881452560425,
917
+ "eval_runtime": 0.9084,
918
+ "eval_samples_per_second": 5.504,
919
+ "eval_steps_per_second": 1.101,
920
+ "step": 640
921
+ },
922
+ {
923
+ "epoch": 0.9992193598750976,
924
+ "step": 640,
925
+ "total_flos": 8.955537329204756e+17,
926
+ "train_loss": 0.4811625644564629,
927
+ "train_runtime": 4514.48,
928
+ "train_samples_per_second": 4.539,
929
+ "train_steps_per_second": 0.142
930
+ }
931
+ ],
932
+ "logging_steps": 5,
933
+ "max_steps": 640,
934
+ "num_input_tokens_seen": 0,
935
+ "num_train_epochs": 1,
936
+ "save_steps": 100,
937
+ "stateful_callbacks": {
938
+ "TrainerControl": {
939
+ "args": {
940
+ "should_epoch_stop": false,
941
+ "should_evaluate": false,
942
+ "should_log": false,
943
+ "should_save": true,
944
+ "should_training_stop": true
945
+ },
946
+ "attributes": {}
947
+ }
948
+ },
949
+ "total_flos": 8.955537329204756e+17,
950
+ "train_batch_size": 2,
951
+ "trial_name": null,
952
+ "trial_params": null
953
+ }