chansung commited on
Commit
f20ac6a
·
verified ·
1 Parent(s): 7299a9c

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_closedqa-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_closedqa-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: 6.9495
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
+ | 5.1634 | 0.9991 | 582 | 6.9495 |
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": 0.9991416309012876,
3
+ "total_flos": 30438567444480.0,
4
+ "train_loss": 6.936546556728402,
5
+ "train_runtime": 8521.4757,
6
+ "train_samples": 111440,
7
+ "train_samples_per_second": 2.186,
8
+ "train_steps_per_second": 0.068
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": 0.9991416309012876,
3
+ "total_flos": 30438567444480.0,
4
+ "train_loss": 6.936546556728402,
5
+ "train_runtime": 8521.4757,
6
+ "train_samples": 111440,
7
+ "train_samples_per_second": 2.186,
8
+ "train_steps_per_second": 0.068
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,869 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.9991416309012876,
5
+ "eval_steps": 500,
6
+ "global_step": 582,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0017167381974248926,
13
+ "grad_norm": 148.37953017406076,
14
+ "learning_rate": 3.3898305084745763e-06,
15
+ "loss": 1.4925,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.008583690987124463,
20
+ "grad_norm": 226.53531287893435,
21
+ "learning_rate": 1.694915254237288e-05,
22
+ "loss": 1.8304,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.017167381974248927,
27
+ "grad_norm": 122.60255331663019,
28
+ "learning_rate": 3.389830508474576e-05,
29
+ "loss": 1.6337,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 0.02575107296137339,
34
+ "grad_norm": 27.57943216063738,
35
+ "learning_rate": 5.0847457627118643e-05,
36
+ "loss": 1.4564,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.034334763948497854,
41
+ "grad_norm": 68.31876381825991,
42
+ "learning_rate": 6.779661016949152e-05,
43
+ "loss": 1.6491,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 0.04291845493562232,
48
+ "grad_norm": 34.24629194904229,
49
+ "learning_rate": 8.474576271186441e-05,
50
+ "loss": 1.6714,
51
+ "step": 25
52
+ },
53
+ {
54
+ "epoch": 0.05150214592274678,
55
+ "grad_norm": 392.0670273900377,
56
+ "learning_rate": 0.00010169491525423729,
57
+ "loss": 2.8097,
58
+ "step": 30
59
+ },
60
+ {
61
+ "epoch": 0.060085836909871244,
62
+ "grad_norm": 1460.2154288260613,
63
+ "learning_rate": 0.00011864406779661017,
64
+ "loss": 2.2112,
65
+ "step": 35
66
+ },
67
+ {
68
+ "epoch": 0.06866952789699571,
69
+ "grad_norm": 152.24800163036767,
70
+ "learning_rate": 0.00013559322033898305,
71
+ "loss": 1.5475,
72
+ "step": 40
73
+ },
74
+ {
75
+ "epoch": 0.07725321888412018,
76
+ "grad_norm": 529.4349457912397,
77
+ "learning_rate": 0.00015254237288135592,
78
+ "loss": 3.2178,
79
+ "step": 45
80
+ },
81
+ {
82
+ "epoch": 0.08583690987124463,
83
+ "grad_norm": 740.2750714071109,
84
+ "learning_rate": 0.00016949152542372882,
85
+ "loss": 3.7609,
86
+ "step": 50
87
+ },
88
+ {
89
+ "epoch": 0.0944206008583691,
90
+ "grad_norm": 501.75647987784066,
91
+ "learning_rate": 0.0001864406779661017,
92
+ "loss": 5.1663,
93
+ "step": 55
94
+ },
95
+ {
96
+ "epoch": 0.10300429184549356,
97
+ "grad_norm": 552.7605592699055,
98
+ "learning_rate": 0.0001999981958814743,
99
+ "loss": 12.807,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 0.11158798283261803,
104
+ "grad_norm": 79.39978511676262,
105
+ "learning_rate": 0.00019993505856798243,
106
+ "loss": 15.5599,
107
+ "step": 65
108
+ },
109
+ {
110
+ "epoch": 0.12017167381974249,
111
+ "grad_norm": 56.78553117050131,
112
+ "learning_rate": 0.00019978178041451998,
113
+ "loss": 16.703,
114
+ "step": 70
115
+ },
116
+ {
117
+ "epoch": 0.12875536480686695,
118
+ "grad_norm": 112.72908285826135,
119
+ "learning_rate": 0.0001995384996770874,
120
+ "loss": 15.425,
121
+ "step": 75
122
+ },
123
+ {
124
+ "epoch": 0.13733905579399142,
125
+ "grad_norm": 442.5710003227928,
126
+ "learning_rate": 0.000199205435793491,
127
+ "loss": 31.9561,
128
+ "step": 80
129
+ },
130
+ {
131
+ "epoch": 0.1459227467811159,
132
+ "grad_norm": 184.54596876342316,
133
+ "learning_rate": 0.00019878288918541114,
134
+ "loss": 15.8809,
135
+ "step": 85
136
+ },
137
+ {
138
+ "epoch": 0.15450643776824036,
139
+ "grad_norm": 347.24694919273804,
140
+ "learning_rate": 0.00019827124098742394,
141
+ "loss": 14.7386,
142
+ "step": 90
143
+ },
144
+ {
145
+ "epoch": 0.1630901287553648,
146
+ "grad_norm": 48.28938306214246,
147
+ "learning_rate": 0.0001976709527032198,
148
+ "loss": 17.4369,
149
+ "step": 95
150
+ },
151
+ {
152
+ "epoch": 0.17167381974248927,
153
+ "grad_norm": 111.82743394628322,
154
+ "learning_rate": 0.00019698256578933005,
155
+ "loss": 13.2218,
156
+ "step": 100
157
+ },
158
+ {
159
+ "epoch": 0.18025751072961374,
160
+ "grad_norm": 266.4352235029355,
161
+ "learning_rate": 0.00019620670116673604,
162
+ "loss": 10.8569,
163
+ "step": 105
164
+ },
165
+ {
166
+ "epoch": 0.1888412017167382,
167
+ "grad_norm": 42.96800119266756,
168
+ "learning_rate": 0.0001953440586608022,
169
+ "loss": 9.8788,
170
+ "step": 110
171
+ },
172
+ {
173
+ "epoch": 0.19742489270386265,
174
+ "grad_norm": 37.647763231857866,
175
+ "learning_rate": 0.00019439541637003772,
176
+ "loss": 8.9958,
177
+ "step": 115
178
+ },
179
+ {
180
+ "epoch": 0.20600858369098712,
181
+ "grad_norm": 20.454288165244996,
182
+ "learning_rate": 0.00019336162996425597,
183
+ "loss": 8.1835,
184
+ "step": 120
185
+ },
186
+ {
187
+ "epoch": 0.2145922746781116,
188
+ "grad_norm": 59.141650986938025,
189
+ "learning_rate": 0.00019224363191276559,
190
+ "loss": 8.2297,
191
+ "step": 125
192
+ },
193
+ {
194
+ "epoch": 0.22317596566523606,
195
+ "grad_norm": 39.88230197372808,
196
+ "learning_rate": 0.00019104243064328838,
197
+ "loss": 8.0509,
198
+ "step": 130
199
+ },
200
+ {
201
+ "epoch": 0.2317596566523605,
202
+ "grad_norm": 27.87553480349785,
203
+ "learning_rate": 0.00018975910963236366,
204
+ "loss": 7.7919,
205
+ "step": 135
206
+ },
207
+ {
208
+ "epoch": 0.24034334763948498,
209
+ "grad_norm": 33.29528319809458,
210
+ "learning_rate": 0.00018839482642805843,
211
+ "loss": 7.4218,
212
+ "step": 140
213
+ },
214
+ {
215
+ "epoch": 0.24892703862660945,
216
+ "grad_norm": 28.706620751101852,
217
+ "learning_rate": 0.000186950811605866,
218
+ "loss": 7.3135,
219
+ "step": 145
220
+ },
221
+ {
222
+ "epoch": 0.2575107296137339,
223
+ "grad_norm": 77.97848200082751,
224
+ "learning_rate": 0.00018542836765873396,
225
+ "loss": 7.3272,
226
+ "step": 150
227
+ },
228
+ {
229
+ "epoch": 0.26609442060085836,
230
+ "grad_norm": 12.617065010078097,
231
+ "learning_rate": 0.00018382886782222313,
232
+ "loss": 7.1705,
233
+ "step": 155
234
+ },
235
+ {
236
+ "epoch": 0.27467811158798283,
237
+ "grad_norm": 9.66255057469759,
238
+ "learning_rate": 0.00018215375483585726,
239
+ "loss": 7.152,
240
+ "step": 160
241
+ },
242
+ {
243
+ "epoch": 0.2832618025751073,
244
+ "grad_norm": 13.81795039749584,
245
+ "learning_rate": 0.0001804045396417801,
246
+ "loss": 7.0384,
247
+ "step": 165
248
+ },
249
+ {
250
+ "epoch": 0.2918454935622318,
251
+ "grad_norm": 10.65607302408338,
252
+ "learning_rate": 0.0001785828000218947,
253
+ "loss": 7.1945,
254
+ "step": 170
255
+ },
256
+ {
257
+ "epoch": 0.30042918454935624,
258
+ "grad_norm": 15.782661016430529,
259
+ "learning_rate": 0.00017669017917471314,
260
+ "loss": 7.7122,
261
+ "step": 175
262
+ },
263
+ {
264
+ "epoch": 0.3090128755364807,
265
+ "grad_norm": 73.67344007196424,
266
+ "learning_rate": 0.00017472838423320108,
267
+ "loss": 7.3337,
268
+ "step": 180
269
+ },
270
+ {
271
+ "epoch": 0.31759656652360513,
272
+ "grad_norm": 14.634220899220917,
273
+ "learning_rate": 0.0001726991847249539,
274
+ "loss": 7.6728,
275
+ "step": 185
276
+ },
277
+ {
278
+ "epoch": 0.3261802575107296,
279
+ "grad_norm": 13.23535439972109,
280
+ "learning_rate": 0.000170604410976093,
281
+ "loss": 7.1979,
282
+ "step": 190
283
+ },
284
+ {
285
+ "epoch": 0.33476394849785407,
286
+ "grad_norm": 15.490178591732896,
287
+ "learning_rate": 0.00016844595246032234,
288
+ "loss": 7.4718,
289
+ "step": 195
290
+ },
291
+ {
292
+ "epoch": 0.34334763948497854,
293
+ "grad_norm": 14.166388152902138,
294
+ "learning_rate": 0.00016622575609463432,
295
+ "loss": 7.0926,
296
+ "step": 200
297
+ },
298
+ {
299
+ "epoch": 0.351931330472103,
300
+ "grad_norm": 7.921484195688758,
301
+ "learning_rate": 0.00016394582448320163,
302
+ "loss": 6.9988,
303
+ "step": 205
304
+ },
305
+ {
306
+ "epoch": 0.3605150214592275,
307
+ "grad_norm": 9.205119757022215,
308
+ "learning_rate": 0.00016160821411104057,
309
+ "loss": 6.9226,
310
+ "step": 210
311
+ },
312
+ {
313
+ "epoch": 0.36909871244635195,
314
+ "grad_norm": 35.99180735659732,
315
+ "learning_rate": 0.00015921503348907288,
316
+ "loss": 7.1323,
317
+ "step": 215
318
+ },
319
+ {
320
+ "epoch": 0.3776824034334764,
321
+ "grad_norm": 11.336602821015093,
322
+ "learning_rate": 0.00015676844125226178,
323
+ "loss": 7.1202,
324
+ "step": 220
325
+ },
326
+ {
327
+ "epoch": 0.38626609442060084,
328
+ "grad_norm": 13.20625274633718,
329
+ "learning_rate": 0.00015427064421253542,
330
+ "loss": 7.2282,
331
+ "step": 225
332
+ },
333
+ {
334
+ "epoch": 0.3948497854077253,
335
+ "grad_norm": 67.29081123770837,
336
+ "learning_rate": 0.00015172389536825554,
337
+ "loss": 7.0692,
338
+ "step": 230
339
+ },
340
+ {
341
+ "epoch": 0.4034334763948498,
342
+ "grad_norm": 11.16921203935729,
343
+ "learning_rate": 0.00014913049187202613,
344
+ "loss": 7.0129,
345
+ "step": 235
346
+ },
347
+ {
348
+ "epoch": 0.41201716738197425,
349
+ "grad_norm": 11.356355094966165,
350
+ "learning_rate": 0.0001464927729586751,
351
+ "loss": 6.9715,
352
+ "step": 240
353
+ },
354
+ {
355
+ "epoch": 0.4206008583690987,
356
+ "grad_norm": 65.02533863066004,
357
+ "learning_rate": 0.0001438131178352782,
358
+ "loss": 7.0428,
359
+ "step": 245
360
+ },
361
+ {
362
+ "epoch": 0.4291845493562232,
363
+ "grad_norm": 8.021375841934983,
364
+ "learning_rate": 0.000141093943535128,
365
+ "loss": 6.9865,
366
+ "step": 250
367
+ },
368
+ {
369
+ "epoch": 0.43776824034334766,
370
+ "grad_norm": 8.30393736182492,
371
+ "learning_rate": 0.00013833770273758396,
372
+ "loss": 6.9422,
373
+ "step": 255
374
+ },
375
+ {
376
+ "epoch": 0.44635193133047213,
377
+ "grad_norm": 10.60254104247866,
378
+ "learning_rate": 0.00013554688155577012,
379
+ "loss": 7.0097,
380
+ "step": 260
381
+ },
382
+ {
383
+ "epoch": 0.45493562231759654,
384
+ "grad_norm": 31.605073666129307,
385
+ "learning_rate": 0.0001327239972941153,
386
+ "loss": 6.9863,
387
+ "step": 265
388
+ },
389
+ {
390
+ "epoch": 0.463519313304721,
391
+ "grad_norm": 8.23597721842936,
392
+ "learning_rate": 0.00012987159617775952,
393
+ "loss": 6.9908,
394
+ "step": 270
395
+ },
396
+ {
397
+ "epoch": 0.4721030042918455,
398
+ "grad_norm": 9.868400148345003,
399
+ "learning_rate": 0.0001269922510558735,
400
+ "loss": 6.9117,
401
+ "step": 275
402
+ },
403
+ {
404
+ "epoch": 0.48068669527896996,
405
+ "grad_norm": 5.852530780039747,
406
+ "learning_rate": 0.00012408855908096372,
407
+ "loss": 6.9009,
408
+ "step": 280
409
+ },
410
+ {
411
+ "epoch": 0.4892703862660944,
412
+ "grad_norm": 5.3374729678346275,
413
+ "learning_rate": 0.00012116313936625616,
414
+ "loss": 6.8528,
415
+ "step": 285
416
+ },
417
+ {
418
+ "epoch": 0.4978540772532189,
419
+ "grad_norm": 5.313298088617677,
420
+ "learning_rate": 0.00011821863062327142,
421
+ "loss": 6.8808,
422
+ "step": 290
423
+ },
424
+ {
425
+ "epoch": 0.5064377682403434,
426
+ "grad_norm": 4.656345284326436,
427
+ "learning_rate": 0.00011525768878172217,
428
+ "loss": 6.8471,
429
+ "step": 295
430
+ },
431
+ {
432
+ "epoch": 0.5150214592274678,
433
+ "grad_norm": 6.2217899411236415,
434
+ "learning_rate": 0.0001122829845938802,
435
+ "loss": 6.8989,
436
+ "step": 300
437
+ },
438
+ {
439
+ "epoch": 0.5236051502145923,
440
+ "grad_norm": 5.068468710318732,
441
+ "learning_rate": 0.00010929720122557346,
442
+ "loss": 6.8636,
443
+ "step": 305
444
+ },
445
+ {
446
+ "epoch": 0.5321888412017167,
447
+ "grad_norm": 4.8062734001678304,
448
+ "learning_rate": 0.00010630303183598586,
449
+ "loss": 6.8643,
450
+ "step": 310
451
+ },
452
+ {
453
+ "epoch": 0.5407725321888412,
454
+ "grad_norm": 4.222058731948505,
455
+ "learning_rate": 0.00010330317714844382,
456
+ "loss": 6.8366,
457
+ "step": 315
458
+ },
459
+ {
460
+ "epoch": 0.5493562231759657,
461
+ "grad_norm": 5.498653200897768,
462
+ "learning_rate": 0.0001003003430143792,
463
+ "loss": 6.8438,
464
+ "step": 320
465
+ },
466
+ {
467
+ "epoch": 0.5579399141630901,
468
+ "grad_norm": 4.606293074529594,
469
+ "learning_rate": 9.729723797266733e-05,
470
+ "loss": 6.8435,
471
+ "step": 325
472
+ },
473
+ {
474
+ "epoch": 0.5665236051502146,
475
+ "grad_norm": 4.094114687291011,
476
+ "learning_rate": 9.429657080654066e-05,
477
+ "loss": 6.8383,
478
+ "step": 330
479
+ },
480
+ {
481
+ "epoch": 0.575107296137339,
482
+ "grad_norm": 4.850653416958559,
483
+ "learning_rate": 9.130104810028196e-05,
484
+ "loss": 6.8403,
485
+ "step": 335
486
+ },
487
+ {
488
+ "epoch": 0.5836909871244635,
489
+ "grad_norm": 3.5962933581788943,
490
+ "learning_rate": 8.831337179790135e-05,
491
+ "loss": 6.7696,
492
+ "step": 340
493
+ },
494
+ {
495
+ "epoch": 0.592274678111588,
496
+ "grad_norm": 4.211863063848454,
497
+ "learning_rate": 8.533623676599832e-05,
498
+ "loss": 6.7095,
499
+ "step": 345
500
+ },
501
+ {
502
+ "epoch": 0.6008583690987125,
503
+ "grad_norm": 2.860326437620046,
504
+ "learning_rate": 8.237232836300797e-05,
505
+ "loss": 6.6912,
506
+ "step": 350
507
+ },
508
+ {
509
+ "epoch": 0.6094420600858369,
510
+ "grad_norm": 8.77242613923949,
511
+ "learning_rate": 7.94243200170232e-05,
512
+ "loss": 6.5853,
513
+ "step": 355
514
+ },
515
+ {
516
+ "epoch": 0.6180257510729614,
517
+ "grad_norm": 6.659382197081032,
518
+ "learning_rate": 7.649487081437818e-05,
519
+ "loss": 6.5642,
520
+ "step": 360
521
+ },
522
+ {
523
+ "epoch": 0.6266094420600858,
524
+ "grad_norm": 10.44433874556097,
525
+ "learning_rate": 7.358662310116808e-05,
526
+ "loss": 6.4968,
527
+ "step": 365
528
+ },
529
+ {
530
+ "epoch": 0.6351931330472103,
531
+ "grad_norm": 6.1233526664383024,
532
+ "learning_rate": 7.070220009986828e-05,
533
+ "loss": 6.4142,
534
+ "step": 370
535
+ },
536
+ {
537
+ "epoch": 0.6437768240343348,
538
+ "grad_norm": 4.045999079298666,
539
+ "learning_rate": 6.784420354320295e-05,
540
+ "loss": 6.3996,
541
+ "step": 375
542
+ },
543
+ {
544
+ "epoch": 0.6523605150214592,
545
+ "grad_norm": 12.756102691389257,
546
+ "learning_rate": 6.501521132739727e-05,
547
+ "loss": 6.3766,
548
+ "step": 380
549
+ },
550
+ {
551
+ "epoch": 0.6609442060085837,
552
+ "grad_norm": 5.067389209546529,
553
+ "learning_rate": 6.221777518693023e-05,
554
+ "loss": 6.3331,
555
+ "step": 385
556
+ },
557
+ {
558
+ "epoch": 0.6695278969957081,
559
+ "grad_norm": 4.587232017547203,
560
+ "learning_rate": 5.945441839288487e-05,
561
+ "loss": 6.2912,
562
+ "step": 390
563
+ },
564
+ {
565
+ "epoch": 0.6781115879828327,
566
+ "grad_norm": 10.154786306342054,
567
+ "learning_rate": 5.672763347697252e-05,
568
+ "loss": 6.1986,
569
+ "step": 395
570
+ },
571
+ {
572
+ "epoch": 0.6866952789699571,
573
+ "grad_norm": 6.356244711188396,
574
+ "learning_rate": 5.403987998328364e-05,
575
+ "loss": 6.0784,
576
+ "step": 400
577
+ },
578
+ {
579
+ "epoch": 0.6952789699570815,
580
+ "grad_norm": 6.256029497303041,
581
+ "learning_rate": 5.139358224979368e-05,
582
+ "loss": 6.0875,
583
+ "step": 405
584
+ },
585
+ {
586
+ "epoch": 0.703862660944206,
587
+ "grad_norm": 6.392716409081631,
588
+ "learning_rate": 4.8791127221624e-05,
589
+ "loss": 6.0507,
590
+ "step": 410
591
+ },
592
+ {
593
+ "epoch": 0.7124463519313304,
594
+ "grad_norm": 4.093689165851618,
595
+ "learning_rate": 4.62348622980316e-05,
596
+ "loss": 6.0207,
597
+ "step": 415
598
+ },
599
+ {
600
+ "epoch": 0.721030042918455,
601
+ "grad_norm": 6.227286133335013,
602
+ "learning_rate": 4.372709321506848e-05,
603
+ "loss": 5.9235,
604
+ "step": 420
605
+ },
606
+ {
607
+ "epoch": 0.7296137339055794,
608
+ "grad_norm": 9.976130870929332,
609
+ "learning_rate": 4.12700819658212e-05,
610
+ "loss": 5.8749,
611
+ "step": 425
612
+ },
613
+ {
614
+ "epoch": 0.7381974248927039,
615
+ "grad_norm": 4.604304131370468,
616
+ "learning_rate": 3.8866044760106446e-05,
617
+ "loss": 5.817,
618
+ "step": 430
619
+ },
620
+ {
621
+ "epoch": 0.7467811158798283,
622
+ "grad_norm": 8.368202258449207,
623
+ "learning_rate": 3.651715002546251e-05,
624
+ "loss": 5.7823,
625
+ "step": 435
626
+ },
627
+ {
628
+ "epoch": 0.7553648068669528,
629
+ "grad_norm": 4.346337001686645,
630
+ "learning_rate": 3.422551645124048e-05,
631
+ "loss": 5.7372,
632
+ "step": 440
633
+ },
634
+ {
635
+ "epoch": 0.7639484978540773,
636
+ "grad_norm": 6.8350119997138306,
637
+ "learning_rate": 3.199321107755863e-05,
638
+ "loss": 5.6973,
639
+ "step": 445
640
+ },
641
+ {
642
+ "epoch": 0.7725321888412017,
643
+ "grad_norm": 4.355909578299472,
644
+ "learning_rate": 2.9822247430844375e-05,
645
+ "loss": 5.6349,
646
+ "step": 450
647
+ },
648
+ {
649
+ "epoch": 0.7811158798283262,
650
+ "grad_norm": 6.751775629287182,
651
+ "learning_rate": 2.7714583707645247e-05,
652
+ "loss": 5.5923,
653
+ "step": 455
654
+ },
655
+ {
656
+ "epoch": 0.7896995708154506,
657
+ "grad_norm": 7.749784220705869,
658
+ "learning_rate": 2.5672121008346895e-05,
659
+ "loss": 5.5167,
660
+ "step": 460
661
+ },
662
+ {
663
+ "epoch": 0.7982832618025751,
664
+ "grad_norm": 10.122654499906968,
665
+ "learning_rate": 2.369670162239166e-05,
666
+ "loss": 5.5233,
667
+ "step": 465
668
+ },
669
+ {
670
+ "epoch": 0.8068669527896996,
671
+ "grad_norm": 13.541428912986023,
672
+ "learning_rate": 2.1790107366544366e-05,
673
+ "loss": 5.4574,
674
+ "step": 470
675
+ },
676
+ {
677
+ "epoch": 0.8154506437768241,
678
+ "grad_norm": 8.344136937270687,
679
+ "learning_rate": 1.9954057977703776e-05,
680
+ "loss": 5.4293,
681
+ "step": 475
682
+ },
683
+ {
684
+ "epoch": 0.8240343347639485,
685
+ "grad_norm": 10.023341035964107,
686
+ "learning_rate": 1.819020956171017e-05,
687
+ "loss": 5.4132,
688
+ "step": 480
689
+ },
690
+ {
691
+ "epoch": 0.8326180257510729,
692
+ "grad_norm": 8.637492625567376,
693
+ "learning_rate": 1.6500153099547344e-05,
694
+ "loss": 5.3985,
695
+ "step": 485
696
+ },
697
+ {
698
+ "epoch": 0.8412017167381974,
699
+ "grad_norm": 7.593060990876112,
700
+ "learning_rate": 1.4885413012287098e-05,
701
+ "loss": 5.3591,
702
+ "step": 490
703
+ },
704
+ {
705
+ "epoch": 0.8497854077253219,
706
+ "grad_norm": 7.293323409290838,
707
+ "learning_rate": 1.3347445786070456e-05,
708
+ "loss": 5.3058,
709
+ "step": 495
710
+ },
711
+ {
712
+ "epoch": 0.8583690987124464,
713
+ "grad_norm": 8.58872146900946,
714
+ "learning_rate": 1.1887638658365452e-05,
715
+ "loss": 5.2812,
716
+ "step": 500
717
+ },
718
+ {
719
+ "epoch": 0.8669527896995708,
720
+ "grad_norm": 10.579607337148637,
721
+ "learning_rate": 1.0507308366687141e-05,
722
+ "loss": 5.2935,
723
+ "step": 505
724
+ },
725
+ {
726
+ "epoch": 0.8755364806866953,
727
+ "grad_norm": 7.796765987218623,
728
+ "learning_rate": 9.207699960908012e-06,
729
+ "loss": 5.3239,
730
+ "step": 510
731
+ },
732
+ {
733
+ "epoch": 0.8841201716738197,
734
+ "grad_norm": 3.845171755230648,
735
+ "learning_rate": 7.989985680230305e-06,
736
+ "loss": 5.284,
737
+ "step": 515
738
+ },
739
+ {
740
+ "epoch": 0.8927038626609443,
741
+ "grad_norm": 3.8997169411441335,
742
+ "learning_rate": 6.855263895833064e-06,
743
+ "loss": 5.2478,
744
+ "step": 520
745
+ },
746
+ {
747
+ "epoch": 0.9012875536480687,
748
+ "grad_norm": 5.516395205208524,
749
+ "learning_rate": 5.804558120147674e-06,
750
+ "loss": 5.2356,
751
+ "step": 525
752
+ },
753
+ {
754
+ "epoch": 0.9098712446351931,
755
+ "grad_norm": 5.160825401670845,
756
+ "learning_rate": 4.838816083655695e-06,
757
+ "loss": 5.2529,
758
+ "step": 530
759
+ },
760
+ {
761
+ "epoch": 0.9184549356223176,
762
+ "grad_norm": 4.170530336706607,
763
+ "learning_rate": 3.95890888004149e-06,
764
+ "loss": 5.2019,
765
+ "step": 535
766
+ },
767
+ {
768
+ "epoch": 0.927038626609442,
769
+ "grad_norm": 4.729325573447989,
770
+ "learning_rate": 3.165630180470802e-06,
771
+ "loss": 5.1868,
772
+ "step": 540
773
+ },
774
+ {
775
+ "epoch": 0.9356223175965666,
776
+ "grad_norm": 4.8268274620539655,
777
+ "learning_rate": 2.459695517703919e-06,
778
+ "loss": 5.194,
779
+ "step": 545
780
+ },
781
+ {
782
+ "epoch": 0.944206008583691,
783
+ "grad_norm": 3.373642611401382,
784
+ "learning_rate": 1.8417416406894094e-06,
785
+ "loss": 5.1408,
786
+ "step": 550
787
+ },
788
+ {
789
+ "epoch": 0.9527896995708155,
790
+ "grad_norm": 2.9560576217055554,
791
+ "learning_rate": 1.3123259402202048e-06,
792
+ "loss": 5.1733,
793
+ "step": 555
794
+ },
795
+ {
796
+ "epoch": 0.9613733905579399,
797
+ "grad_norm": 3.364514157359552,
798
+ "learning_rate": 8.71925946170371e-07,
799
+ "loss": 5.1221,
800
+ "step": 560
801
+ },
802
+ {
803
+ "epoch": 0.9699570815450643,
804
+ "grad_norm": 2.8157940718449845,
805
+ "learning_rate": 5.209388967659768e-07,
806
+ "loss": 5.1479,
807
+ "step": 565
808
+ },
809
+ {
810
+ "epoch": 0.9785407725321889,
811
+ "grad_norm": 3.778011781318662,
812
+ "learning_rate": 2.596813802785092e-07,
813
+ "loss": 5.1574,
814
+ "step": 570
815
+ },
816
+ {
817
+ "epoch": 0.9871244635193133,
818
+ "grad_norm": 3.0834295680511814,
819
+ "learning_rate": 8.838904946407711e-08,
820
+ "loss": 5.1232,
821
+ "step": 575
822
+ },
823
+ {
824
+ "epoch": 0.9957081545064378,
825
+ "grad_norm": 3.298741897550586,
826
+ "learning_rate": 7.216409006038927e-09,
827
+ "loss": 5.1634,
828
+ "step": 580
829
+ },
830
+ {
831
+ "epoch": 0.9991416309012876,
832
+ "eval_loss": 6.949496746063232,
833
+ "eval_runtime": 2.3831,
834
+ "eval_samples_per_second": 2.518,
835
+ "eval_steps_per_second": 0.42,
836
+ "step": 582
837
+ },
838
+ {
839
+ "epoch": 0.9991416309012876,
840
+ "step": 582,
841
+ "total_flos": 30438567444480.0,
842
+ "train_loss": 6.936546556728402,
843
+ "train_runtime": 8521.4757,
844
+ "train_samples_per_second": 2.186,
845
+ "train_steps_per_second": 0.068
846
+ }
847
+ ],
848
+ "logging_steps": 5,
849
+ "max_steps": 582,
850
+ "num_input_tokens_seen": 0,
851
+ "num_train_epochs": 1,
852
+ "save_steps": 100,
853
+ "stateful_callbacks": {
854
+ "TrainerControl": {
855
+ "args": {
856
+ "should_epoch_stop": false,
857
+ "should_evaluate": false,
858
+ "should_log": false,
859
+ "should_save": true,
860
+ "should_training_stop": true
861
+ },
862
+ "attributes": {}
863
+ }
864
+ },
865
+ "total_flos": 30438567444480.0,
866
+ "train_batch_size": 2,
867
+ "trial_name": null,
868
+ "trial_params": null
869
+ }