chansung commited on
Commit
6f3318e
·
verified ·
1 Parent(s): 732c1ea

Model save

Browse files
README.md ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: llama3
4
+ base_model: meta-llama/Meta-Llama-3-8B
5
+ tags:
6
+ - trl
7
+ - sft
8
+ - generated_from_trainer
9
+ datasets:
10
+ - generator
11
+ model-index:
12
+ - name: llama3.1-8b-gpt4o_100k_coding-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
+ # llama3.1-8b-gpt4o_100k_coding-fft
20
+
21
+ This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the generator dataset.
22
+ It achieves the following results on the evaluation set:
23
+ - Loss: 2.2535
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: 0.0003
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
+ | 0.5499 | 0.9991 | 539 | 2.2535 |
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": 0.9990732159406858,
3
+ "total_flos": 28187736145920.0,
4
+ "train_loss": 1.0674916249259283,
5
+ "train_runtime": 10422.8572,
6
+ "train_samples": 116368,
7
+ "train_samples_per_second": 1.655,
8
+ "train_steps_per_second": 0.052
9
+ }
generation_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 128256,
3
+ "do_sample": true,
4
+ "eos_token_id": 128257,
5
+ "max_length": 4096,
6
+ "pad_token_id": 128257,
7
+ "temperature": 0.6,
8
+ "top_p": 0.9,
9
+ "transformers_version": "4.45.1"
10
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 0.9990732159406858,
3
+ "total_flos": 28187736145920.0,
4
+ "train_loss": 1.0674916249259283,
5
+ "train_runtime": 10422.8572,
6
+ "train_samples": 116368,
7
+ "train_samples_per_second": 1.655,
8
+ "train_steps_per_second": 0.052
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,806 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.9990732159406858,
5
+ "eval_steps": 500,
6
+ "global_step": 539,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0018535681186283596,
13
+ "grad_norm": 26.35410018433926,
14
+ "learning_rate": 5.555555555555555e-06,
15
+ "loss": 1.6809,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.009267840593141797,
20
+ "grad_norm": 39.94533926724377,
21
+ "learning_rate": 2.7777777777777772e-05,
22
+ "loss": 1.343,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.018535681186283594,
27
+ "grad_norm": 4.5543620999927485,
28
+ "learning_rate": 5.5555555555555545e-05,
29
+ "loss": 0.8749,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 0.027803521779425393,
34
+ "grad_norm": 4.24495034141875,
35
+ "learning_rate": 8.333333333333333e-05,
36
+ "loss": 0.8704,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.03707136237256719,
41
+ "grad_norm": 39.42543160126464,
42
+ "learning_rate": 0.00011111111111111109,
43
+ "loss": 0.8733,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 0.04633920296570899,
48
+ "grad_norm": 11.962483325702362,
49
+ "learning_rate": 0.0001388888888888889,
50
+ "loss": 1.0553,
51
+ "step": 25
52
+ },
53
+ {
54
+ "epoch": 0.05560704355885079,
55
+ "grad_norm": 14.195688905715766,
56
+ "learning_rate": 0.00016666666666666666,
57
+ "loss": 1.0623,
58
+ "step": 30
59
+ },
60
+ {
61
+ "epoch": 0.06487488415199258,
62
+ "grad_norm": 4.164105779251046,
63
+ "learning_rate": 0.00019444444444444443,
64
+ "loss": 0.8718,
65
+ "step": 35
66
+ },
67
+ {
68
+ "epoch": 0.07414272474513438,
69
+ "grad_norm": 6.991113945747281,
70
+ "learning_rate": 0.00022222222222222218,
71
+ "loss": 0.8619,
72
+ "step": 40
73
+ },
74
+ {
75
+ "epoch": 0.08341056533827618,
76
+ "grad_norm": 631.4969668103296,
77
+ "learning_rate": 0.00025,
78
+ "loss": 1.8292,
79
+ "step": 45
80
+ },
81
+ {
82
+ "epoch": 0.09267840593141798,
83
+ "grad_norm": 51.99035558508526,
84
+ "learning_rate": 0.0002777777777777778,
85
+ "loss": 1.6128,
86
+ "step": 50
87
+ },
88
+ {
89
+ "epoch": 0.10194624652455977,
90
+ "grad_norm": 5.615188207831111,
91
+ "learning_rate": 0.0002999968531502098,
92
+ "loss": 1.1458,
93
+ "step": 55
94
+ },
95
+ {
96
+ "epoch": 0.11121408711770157,
97
+ "grad_norm": 162.57564664314097,
98
+ "learning_rate": 0.0002998867272706619,
99
+ "loss": 1.7169,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 0.12048192771084337,
104
+ "grad_norm": 27.819505339583905,
105
+ "learning_rate": 0.0002996193909122197,
106
+ "loss": 2.1196,
107
+ "step": 65
108
+ },
109
+ {
110
+ "epoch": 0.12974976830398516,
111
+ "grad_norm": 12.017979123334465,
112
+ "learning_rate": 0.00029919512447380625,
113
+ "loss": 1.3348,
114
+ "step": 70
115
+ },
116
+ {
117
+ "epoch": 0.13901760889712697,
118
+ "grad_norm": 529.5307829885717,
119
+ "learning_rate": 0.0002986143729523282,
120
+ "loss": 1.0105,
121
+ "step": 75
122
+ },
123
+ {
124
+ "epoch": 0.14828544949026876,
125
+ "grad_norm": 2.511805320667406,
126
+ "learning_rate": 0.000297877745475935,
127
+ "loss": 1.0586,
128
+ "step": 80
129
+ },
130
+ {
131
+ "epoch": 0.15755329008341057,
132
+ "grad_norm": 4.022429995067504,
133
+ "learning_rate": 0.0002969860146651276,
134
+ "loss": 0.9055,
135
+ "step": 85
136
+ },
137
+ {
138
+ "epoch": 0.16682113067655235,
139
+ "grad_norm": 512.3998660297882,
140
+ "learning_rate": 0.0002959401158223867,
141
+ "loss": 5.3655,
142
+ "step": 90
143
+ },
144
+ {
145
+ "epoch": 0.17608897126969417,
146
+ "grad_norm": 131.99417407337796,
147
+ "learning_rate": 0.00029474114595116896,
148
+ "loss": 3.1238,
149
+ "step": 95
150
+ },
151
+ {
152
+ "epoch": 0.18535681186283595,
153
+ "grad_norm": 19.953838142892188,
154
+ "learning_rate": 0.0002933903626053024,
155
+ "loss": 1.9603,
156
+ "step": 100
157
+ },
158
+ {
159
+ "epoch": 0.19462465245597776,
160
+ "grad_norm": 7.472797510255509,
161
+ "learning_rate": 0.00029188918256998564,
162
+ "loss": 1.4801,
163
+ "step": 105
164
+ },
165
+ {
166
+ "epoch": 0.20389249304911955,
167
+ "grad_norm": 9.70552960980865,
168
+ "learning_rate": 0.00029023918037577635,
169
+ "loss": 1.3544,
170
+ "step": 110
171
+ },
172
+ {
173
+ "epoch": 0.21316033364226136,
174
+ "grad_norm": 17.11916105312361,
175
+ "learning_rate": 0.00028844208664712575,
176
+ "loss": 1.0681,
177
+ "step": 115
178
+ },
179
+ {
180
+ "epoch": 0.22242817423540315,
181
+ "grad_norm": 22.28855318123011,
182
+ "learning_rate": 0.00028649978628719254,
183
+ "loss": 1.2611,
184
+ "step": 120
185
+ },
186
+ {
187
+ "epoch": 0.23169601482854496,
188
+ "grad_norm": 34.76579789562846,
189
+ "learning_rate": 0.00028441431650084016,
190
+ "loss": 1.6181,
191
+ "step": 125
192
+ },
193
+ {
194
+ "epoch": 0.24096385542168675,
195
+ "grad_norm": 15.11803328233237,
196
+ "learning_rate": 0.0002821878646578898,
197
+ "loss": 1.1601,
198
+ "step": 130
199
+ },
200
+ {
201
+ "epoch": 0.25023169601482853,
202
+ "grad_norm": 10.594239787348334,
203
+ "learning_rate": 0.0002798227659988717,
204
+ "loss": 1.0309,
205
+ "step": 135
206
+ },
207
+ {
208
+ "epoch": 0.2594995366079703,
209
+ "grad_norm": 3.881883182518721,
210
+ "learning_rate": 0.00027732150118568017,
211
+ "loss": 1.5651,
212
+ "step": 140
213
+ },
214
+ {
215
+ "epoch": 0.26876737720111216,
216
+ "grad_norm": 3.036719624432077,
217
+ "learning_rate": 0.00027468669369970207,
218
+ "loss": 1.1445,
219
+ "step": 145
220
+ },
221
+ {
222
+ "epoch": 0.27803521779425394,
223
+ "grad_norm": 14.352629655607771,
224
+ "learning_rate": 0.00027192110709014697,
225
+ "loss": 0.9305,
226
+ "step": 150
227
+ },
228
+ {
229
+ "epoch": 0.2873030583873957,
230
+ "grad_norm": 2.933168446331413,
231
+ "learning_rate": 0.0002690276420754655,
232
+ "loss": 0.9324,
233
+ "step": 155
234
+ },
235
+ {
236
+ "epoch": 0.2965708989805375,
237
+ "grad_norm": 4.661928504358525,
238
+ "learning_rate": 0.00026600933350089654,
239
+ "loss": 0.9491,
240
+ "step": 160
241
+ },
242
+ {
243
+ "epoch": 0.30583873957367935,
244
+ "grad_norm": 5.782107803222577,
245
+ "learning_rate": 0.0002628693471553335,
246
+ "loss": 0.8689,
247
+ "step": 165
248
+ },
249
+ {
250
+ "epoch": 0.31510658016682114,
251
+ "grad_norm": 4.287118434199453,
252
+ "learning_rate": 0.00025961097645084885,
253
+ "loss": 1.112,
254
+ "step": 170
255
+ },
256
+ {
257
+ "epoch": 0.3243744207599629,
258
+ "grad_norm": 170.8246279898043,
259
+ "learning_rate": 0.0002562376389683599,
260
+ "loss": 2.2669,
261
+ "step": 175
262
+ },
263
+ {
264
+ "epoch": 0.3336422613531047,
265
+ "grad_norm": 243.40178580373365,
266
+ "learning_rate": 0.00025275287287305814,
267
+ "loss": 3.2917,
268
+ "step": 180
269
+ },
270
+ {
271
+ "epoch": 0.34291010194624655,
272
+ "grad_norm": 160.38147120987801,
273
+ "learning_rate": 0.00024916033320336264,
274
+ "loss": 3.067,
275
+ "step": 185
276
+ },
277
+ {
278
+ "epoch": 0.35217794253938833,
279
+ "grad_norm": 30.070071532273786,
280
+ "learning_rate": 0.0002454637880372892,
281
+ "loss": 3.1301,
282
+ "step": 190
283
+ },
284
+ {
285
+ "epoch": 0.3614457831325301,
286
+ "grad_norm": 48.47525499191394,
287
+ "learning_rate": 0.0002416671145402575,
288
+ "loss": 2.6178,
289
+ "step": 195
290
+ },
291
+ {
292
+ "epoch": 0.3707136237256719,
293
+ "grad_norm": 48.37664262960068,
294
+ "learning_rate": 0.00023777429489847934,
295
+ "loss": 1.441,
296
+ "step": 200
297
+ },
298
+ {
299
+ "epoch": 0.3799814643188137,
300
+ "grad_norm": 3.8665655676681245,
301
+ "learning_rate": 0.0002337894121421954,
302
+ "loss": 1.2978,
303
+ "step": 205
304
+ },
305
+ {
306
+ "epoch": 0.38924930491195553,
307
+ "grad_norm": 2.9273559971935055,
308
+ "learning_rate": 0.00022971664586314054,
309
+ "loss": 1.0238,
310
+ "step": 210
311
+ },
312
+ {
313
+ "epoch": 0.3985171455050973,
314
+ "grad_norm": 20.165413444524287,
315
+ "learning_rate": 0.00022556026783072895,
316
+ "loss": 0.8787,
317
+ "step": 215
318
+ },
319
+ {
320
+ "epoch": 0.4077849860982391,
321
+ "grad_norm": 2.3870488681045203,
322
+ "learning_rate": 0.00022132463751155812,
323
+ "loss": 0.9467,
324
+ "step": 220
325
+ },
326
+ {
327
+ "epoch": 0.4170528266913809,
328
+ "grad_norm": 4.038604860534557,
329
+ "learning_rate": 0.00021701419749693034,
330
+ "loss": 0.9708,
331
+ "step": 225
332
+ },
333
+ {
334
+ "epoch": 0.4263206672845227,
335
+ "grad_norm": 2.7240106071190535,
336
+ "learning_rate": 0.00021263346884318777,
337
+ "loss": 1.0564,
338
+ "step": 230
339
+ },
340
+ {
341
+ "epoch": 0.4355885078776645,
342
+ "grad_norm": 1.6377519083501402,
343
+ "learning_rate": 0.00020818704632974896,
344
+ "loss": 0.7724,
345
+ "step": 235
346
+ },
347
+ {
348
+ "epoch": 0.4448563484708063,
349
+ "grad_norm": 1.0433662853439323,
350
+ "learning_rate": 0.00020367959363981936,
351
+ "loss": 0.8052,
352
+ "step": 240
353
+ },
354
+ {
355
+ "epoch": 0.4541241890639481,
356
+ "grad_norm": 1.5351301360333338,
357
+ "learning_rate": 0.00019911583846883197,
358
+ "loss": 0.8407,
359
+ "step": 245
360
+ },
361
+ {
362
+ "epoch": 0.4633920296570899,
363
+ "grad_norm": 1.2741936551633426,
364
+ "learning_rate": 0.0001945005675657475,
365
+ "loss": 0.8344,
366
+ "step": 250
367
+ },
368
+ {
369
+ "epoch": 0.4726598702502317,
370
+ "grad_norm": 2.4652187839401467,
371
+ "learning_rate": 0.00018983862171241577,
372
+ "loss": 0.9688,
373
+ "step": 255
374
+ },
375
+ {
376
+ "epoch": 0.4819277108433735,
377
+ "grad_norm": 1.1552631433359035,
378
+ "learning_rate": 0.00018513489064626398,
379
+ "loss": 0.8647,
380
+ "step": 260
381
+ },
382
+ {
383
+ "epoch": 0.4911955514365153,
384
+ "grad_norm": 1.113328634852908,
385
+ "learning_rate": 0.00018039430793163753,
386
+ "loss": 0.8514,
387
+ "step": 265
388
+ },
389
+ {
390
+ "epoch": 0.5004633920296571,
391
+ "grad_norm": 0.995245914359209,
392
+ "learning_rate": 0.00017562184578517203,
393
+ "loss": 0.8845,
394
+ "step": 270
395
+ },
396
+ {
397
+ "epoch": 0.5097312326227988,
398
+ "grad_norm": 1.5488835001137233,
399
+ "learning_rate": 0.00017082250986062502,
400
+ "loss": 0.7809,
401
+ "step": 275
402
+ },
403
+ {
404
+ "epoch": 0.5189990732159406,
405
+ "grad_norm": 1.5756484423065427,
406
+ "learning_rate": 0.00016600133399863594,
407
+ "loss": 0.8326,
408
+ "step": 280
409
+ },
410
+ {
411
+ "epoch": 0.5282669138090825,
412
+ "grad_norm": 206.03580929049917,
413
+ "learning_rate": 0.0001611633749469231,
414
+ "loss": 2.2034,
415
+ "step": 285
416
+ },
417
+ {
418
+ "epoch": 0.5375347544022243,
419
+ "grad_norm": 4.169210184616496,
420
+ "learning_rate": 0.0001563137070564528,
421
+ "loss": 1.9396,
422
+ "step": 290
423
+ },
424
+ {
425
+ "epoch": 0.5468025949953661,
426
+ "grad_norm": 7.018651154352153,
427
+ "learning_rate": 0.0001514574169591466,
428
+ "loss": 1.4702,
429
+ "step": 295
430
+ },
431
+ {
432
+ "epoch": 0.5560704355885079,
433
+ "grad_norm": 1.7304196471580864,
434
+ "learning_rate": 0.0001465995982327065,
435
+ "loss": 0.9425,
436
+ "step": 300
437
+ },
438
+ {
439
+ "epoch": 0.5653382761816497,
440
+ "grad_norm": 2.241775914452003,
441
+ "learning_rate": 0.00014174534605815525,
442
+ "loss": 0.7961,
443
+ "step": 305
444
+ },
445
+ {
446
+ "epoch": 0.5746061167747915,
447
+ "grad_norm": 3.456242970621782,
448
+ "learning_rate": 0.0001368997518756954,
449
+ "loss": 0.834,
450
+ "step": 310
451
+ },
452
+ {
453
+ "epoch": 0.5838739573679332,
454
+ "grad_norm": 5.072902047355265,
455
+ "learning_rate": 0.00013206789804449116,
456
+ "loss": 0.99,
457
+ "step": 315
458
+ },
459
+ {
460
+ "epoch": 0.593141797961075,
461
+ "grad_norm": 6.231760889035276,
462
+ "learning_rate": 0.0001272548525119758,
463
+ "loss": 0.8985,
464
+ "step": 320
465
+ },
466
+ {
467
+ "epoch": 0.6024096385542169,
468
+ "grad_norm": 0.8710278075890822,
469
+ "learning_rate": 0.0001224656634982746,
470
+ "loss": 0.7725,
471
+ "step": 325
472
+ },
473
+ {
474
+ "epoch": 0.6116774791473587,
475
+ "grad_norm": 1.799368795567863,
476
+ "learning_rate": 0.00011770535420131876,
477
+ "loss": 0.7621,
478
+ "step": 330
479
+ },
480
+ {
481
+ "epoch": 0.6209453197405005,
482
+ "grad_norm": 0.9742992413928552,
483
+ "learning_rate": 0.00011297891752820484,
484
+ "loss": 0.8327,
485
+ "step": 335
486
+ },
487
+ {
488
+ "epoch": 0.6302131603336423,
489
+ "grad_norm": 0.85009149768552,
490
+ "learning_rate": 0.0001082913108583245,
491
+ "loss": 0.6863,
492
+ "step": 340
493
+ },
494
+ {
495
+ "epoch": 0.6394810009267841,
496
+ "grad_norm": 0.8181844995131707,
497
+ "learning_rate": 0.0001036474508437579,
498
+ "loss": 0.6779,
499
+ "step": 345
500
+ },
501
+ {
502
+ "epoch": 0.6487488415199258,
503
+ "grad_norm": 0.9546399988989202,
504
+ "learning_rate": 9.905220825238491e-05,
505
+ "loss": 0.6872,
506
+ "step": 350
507
+ },
508
+ {
509
+ "epoch": 0.6580166821130676,
510
+ "grad_norm": 0.6995327936772554,
511
+ "learning_rate": 9.45104028591222e-05,
512
+ "loss": 0.7025,
513
+ "step": 355
514
+ },
515
+ {
516
+ "epoch": 0.6672845227062094,
517
+ "grad_norm": 0.7195227820211112,
518
+ "learning_rate": 9.002679839064463e-05,
519
+ "loss": 0.6807,
520
+ "step": 360
521
+ },
522
+ {
523
+ "epoch": 0.6765523632993512,
524
+ "grad_norm": 0.6283846451216366,
525
+ "learning_rate": 8.560609752889412e-05,
526
+ "loss": 0.6888,
527
+ "step": 365
528
+ },
529
+ {
530
+ "epoch": 0.6858202038924931,
531
+ "grad_norm": 0.7394400936445648,
532
+ "learning_rate": 8.125293697861548e-05,
533
+ "loss": 0.6542,
534
+ "step": 370
535
+ },
536
+ {
537
+ "epoch": 0.6950880444856349,
538
+ "grad_norm": 0.6824620929581083,
539
+ "learning_rate": 7.697188260409356e-05,
540
+ "loss": 0.671,
541
+ "step": 375
542
+ },
543
+ {
544
+ "epoch": 0.7043558850787767,
545
+ "grad_norm": 0.6116075298871171,
546
+ "learning_rate": 7.276742464019198e-05,
547
+ "loss": 0.6729,
548
+ "step": 380
549
+ },
550
+ {
551
+ "epoch": 0.7136237256719185,
552
+ "grad_norm": 0.6453928151826652,
553
+ "learning_rate": 6.864397298271699e-05,
554
+ "loss": 0.6626,
555
+ "step": 385
556
+ },
557
+ {
558
+ "epoch": 0.7228915662650602,
559
+ "grad_norm": 0.6716078881868212,
560
+ "learning_rate": 6.460585256304559e-05,
561
+ "loss": 0.6851,
562
+ "step": 390
563
+ },
564
+ {
565
+ "epoch": 0.732159406858202,
566
+ "grad_norm": 0.6801638199214254,
567
+ "learning_rate": 6.065729881186982e-05,
568
+ "loss": 0.6168,
569
+ "step": 395
570
+ },
571
+ {
572
+ "epoch": 0.7414272474513438,
573
+ "grad_norm": 0.4794662188577268,
574
+ "learning_rate": 5.680245321681471e-05,
575
+ "loss": 0.6531,
576
+ "step": 400
577
+ },
578
+ {
579
+ "epoch": 0.7506950880444856,
580
+ "grad_norm": 1.029961404045439,
581
+ "learning_rate": 5.304535897858999e-05,
582
+ "loss": 0.6295,
583
+ "step": 405
584
+ },
585
+ {
586
+ "epoch": 0.7599629286376274,
587
+ "grad_norm": 0.5284338458030478,
588
+ "learning_rate": 4.938995677023054e-05,
589
+ "loss": 0.6201,
590
+ "step": 410
591
+ },
592
+ {
593
+ "epoch": 0.7692307692307693,
594
+ "grad_norm": 0.4516339145236783,
595
+ "learning_rate": 4.584008060387455e-05,
596
+ "loss": 0.6083,
597
+ "step": 415
598
+ },
599
+ {
600
+ "epoch": 0.7784986098239111,
601
+ "grad_norm": 0.5727589086366411,
602
+ "learning_rate": 4.239945380941461e-05,
603
+ "loss": 0.6021,
604
+ "step": 420
605
+ },
606
+ {
607
+ "epoch": 0.7877664504170528,
608
+ "grad_norm": 0.4804742186453503,
609
+ "learning_rate": 3.907168512923842e-05,
610
+ "loss": 0.5933,
611
+ "step": 425
612
+ },
613
+ {
614
+ "epoch": 0.7970342910101946,
615
+ "grad_norm": 0.45776606255626884,
616
+ "learning_rate": 3.5860264933156324e-05,
617
+ "loss": 0.5774,
618
+ "step": 430
619
+ },
620
+ {
621
+ "epoch": 0.8063021316033364,
622
+ "grad_norm": 1.2302657626291176,
623
+ "learning_rate": 3.276856155748584e-05,
624
+ "loss": 0.5908,
625
+ "step": 435
626
+ },
627
+ {
628
+ "epoch": 0.8155699721964782,
629
+ "grad_norm": 0.9348743887035583,
630
+ "learning_rate": 2.9799817772131516e-05,
631
+ "loss": 0.592,
632
+ "step": 440
633
+ },
634
+ {
635
+ "epoch": 0.82483781278962,
636
+ "grad_norm": 0.4027186043623421,
637
+ "learning_rate": 2.6957147379367217e-05,
638
+ "loss": 0.5798,
639
+ "step": 445
640
+ },
641
+ {
642
+ "epoch": 0.8341056533827618,
643
+ "grad_norm": 0.5240634600695104,
644
+ "learning_rate": 2.4243531947887802e-05,
645
+ "loss": 0.5805,
646
+ "step": 450
647
+ },
648
+ {
649
+ "epoch": 0.8433734939759037,
650
+ "grad_norm": 0.5416112282286268,
651
+ "learning_rate": 2.1661817685554833e-05,
652
+ "loss": 0.6067,
653
+ "step": 455
654
+ },
655
+ {
656
+ "epoch": 0.8526413345690455,
657
+ "grad_norm": 0.47266676591358336,
658
+ "learning_rate": 1.921471245411794e-05,
659
+ "loss": 0.5962,
660
+ "step": 460
661
+ },
662
+ {
663
+ "epoch": 0.8619091751621872,
664
+ "grad_norm": 0.6382596120183166,
665
+ "learning_rate": 1.6904782929041693e-05,
666
+ "loss": 0.5791,
667
+ "step": 465
668
+ },
669
+ {
670
+ "epoch": 0.871177015755329,
671
+ "grad_norm": 0.49306751144827193,
672
+ "learning_rate": 1.4734451907417255e-05,
673
+ "loss": 0.6182,
674
+ "step": 470
675
+ },
676
+ {
677
+ "epoch": 0.8804448563484708,
678
+ "grad_norm": 0.49635162924553927,
679
+ "learning_rate": 1.2705995766783079e-05,
680
+ "loss": 0.5521,
681
+ "step": 475
682
+ },
683
+ {
684
+ "epoch": 0.8897126969416126,
685
+ "grad_norm": 0.49868563464480065,
686
+ "learning_rate": 1.0821542077519169e-05,
687
+ "loss": 0.5579,
688
+ "step": 480
689
+ },
690
+ {
691
+ "epoch": 0.8989805375347544,
692
+ "grad_norm": 0.384294096915848,
693
+ "learning_rate": 9.083067371319324e-06,
694
+ "loss": 0.5532,
695
+ "step": 485
696
+ },
697
+ {
698
+ "epoch": 0.9082483781278962,
699
+ "grad_norm": 0.6584770004045724,
700
+ "learning_rate": 7.492395068082619e-06,
701
+ "loss": 0.544,
702
+ "step": 490
703
+ },
704
+ {
705
+ "epoch": 0.917516218721038,
706
+ "grad_norm": 0.5198655418147264,
707
+ "learning_rate": 6.051193563397599e-06,
708
+ "loss": 0.555,
709
+ "step": 495
710
+ },
711
+ {
712
+ "epoch": 0.9267840593141798,
713
+ "grad_norm": 0.46652880015863485,
714
+ "learning_rate": 4.760974478625634e-06,
715
+ "loss": 0.5437,
716
+ "step": 500
717
+ },
718
+ {
719
+ "epoch": 0.9360518999073216,
720
+ "grad_norm": 1.0569529661255628,
721
+ "learning_rate": 3.623091075418977e-06,
722
+ "loss": 0.5573,
723
+ "step": 505
724
+ },
725
+ {
726
+ "epoch": 0.9453197405004634,
727
+ "grad_norm": 0.4373874071284105,
728
+ "learning_rate": 2.638736836336158e-06,
729
+ "loss": 0.5312,
730
+ "step": 510
731
+ },
732
+ {
733
+ "epoch": 0.9545875810936052,
734
+ "grad_norm": 0.46826046078794514,
735
+ "learning_rate": 1.8089442130434061e-06,
736
+ "loss": 0.5648,
737
+ "step": 515
738
+ },
739
+ {
740
+ "epoch": 0.963855421686747,
741
+ "grad_norm": 0.40638149868456475,
742
+ "learning_rate": 1.1345835434156736e-06,
743
+ "loss": 0.5417,
744
+ "step": 520
745
+ },
746
+ {
747
+ "epoch": 0.9731232622798888,
748
+ "grad_norm": 0.4274920327067933,
749
+ "learning_rate": 6.163621386722218e-07,
750
+ "loss": 0.5528,
751
+ "step": 525
752
+ },
753
+ {
754
+ "epoch": 0.9823911028730306,
755
+ "grad_norm": 0.41954954402917344,
756
+ "learning_rate": 2.5482354150493935e-07,
757
+ "loss": 0.549,
758
+ "step": 530
759
+ },
760
+ {
761
+ "epoch": 0.9916589434661723,
762
+ "grad_norm": 0.46407841167410513,
763
+ "learning_rate": 5.0346955976976467e-08,
764
+ "loss": 0.5499,
765
+ "step": 535
766
+ },
767
+ {
768
+ "epoch": 0.9990732159406858,
769
+ "eval_loss": 2.2535195350646973,
770
+ "eval_runtime": 2.3617,
771
+ "eval_samples_per_second": 1.694,
772
+ "eval_steps_per_second": 0.423,
773
+ "step": 539
774
+ },
775
+ {
776
+ "epoch": 0.9990732159406858,
777
+ "step": 539,
778
+ "total_flos": 28187736145920.0,
779
+ "train_loss": 1.0674916249259283,
780
+ "train_runtime": 10422.8572,
781
+ "train_samples_per_second": 1.655,
782
+ "train_steps_per_second": 0.052
783
+ }
784
+ ],
785
+ "logging_steps": 5,
786
+ "max_steps": 539,
787
+ "num_input_tokens_seen": 0,
788
+ "num_train_epochs": 1,
789
+ "save_steps": 100,
790
+ "stateful_callbacks": {
791
+ "TrainerControl": {
792
+ "args": {
793
+ "should_epoch_stop": false,
794
+ "should_evaluate": false,
795
+ "should_log": false,
796
+ "should_save": true,
797
+ "should_training_stop": true
798
+ },
799
+ "attributes": {}
800
+ }
801
+ },
802
+ "total_flos": 28187736145920.0,
803
+ "train_batch_size": 2,
804
+ "trial_name": null,
805
+ "trial_params": null
806
+ }