flyingbugs commited on
Commit
90f6eaf
·
verified ·
1 Parent(s): bfe61c2

Model save

Browse files
README.md ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: Qwen/Qwen2.5-1.5B-Instruct
3
+ library_name: transformers
4
+ model_name: Qwen2.5-1.5B-Open-R1-Distill-eos-epic
5
+ tags:
6
+ - generated_from_trainer
7
+ - trl
8
+ - sft
9
+ licence: license
10
+ ---
11
+
12
+ # Model Card for Qwen2.5-1.5B-Open-R1-Distill-eos-epic
13
+
14
+ This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct).
15
+ It has been trained using [TRL](https://github.com/huggingface/trl).
16
+
17
+ ## Quick start
18
+
19
+ ```python
20
+ from transformers import pipeline
21
+
22
+ question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
23
+ generator = pipeline("text-generation", model="flyingbugs/Qwen2.5-1.5B-Open-R1-Distill-eos-epic", device="cuda")
24
+ output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
25
+ print(output["generated_text"])
26
+ ```
27
+
28
+ ## Training procedure
29
+
30
+ [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/jjh233/huggingface/runs/371aa1gg)
31
+
32
+
33
+ This model was trained with SFT.
34
+
35
+ ### Framework versions
36
+
37
+ - TRL: 0.16.0.dev0
38
+ - Transformers: 4.54.0
39
+ - Pytorch: 2.5.1
40
+ - Datasets: 4.0.0
41
+ - Tokenizers: 0.21.2
42
+
43
+ ## Citations
44
+
45
+
46
+
47
+ Cite TRL as:
48
+
49
+ ```bibtex
50
+ @misc{vonwerra2022trl,
51
+ title = {{TRL: Transformer Reinforcement Learning}},
52
+ author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
53
+ year = 2020,
54
+ journal = {GitHub repository},
55
+ publisher = {GitHub},
56
+ howpublished = {\url{https://github.com/huggingface/trl}}
57
+ }
58
+ ```
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "total_flos": 962656789856256.0,
3
+ "train_loss": 0.551490586363908,
4
+ "train_runtime": 11763.4703,
5
+ "train_samples": 93733,
6
+ "train_samples_per_second": 5.731,
7
+ "train_steps_per_second": 0.045
8
+ }
generation_config.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 151643,
3
+ "do_sample": true,
4
+ "eos_token_id": [
5
+ 151645,
6
+ 151643
7
+ ],
8
+ "pad_token_id": 151643,
9
+ "repetition_penalty": 1.1,
10
+ "temperature": 0.7,
11
+ "top_k": 20,
12
+ "top_p": 0.8,
13
+ "transformers_version": "4.54.0"
14
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "total_flos": 962656789856256.0,
3
+ "train_loss": 0.551490586363908,
4
+ "train_runtime": 11763.4703,
5
+ "train_samples": 93733,
6
+ "train_samples_per_second": 5.731,
7
+ "train_steps_per_second": 0.045
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,778 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 528,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.028409090909090908,
14
+ "grad_norm": 1.5463851071683217,
15
+ "learning_rate": 7.4074074074074075e-06,
16
+ "loss": 0.8827,
17
+ "step": 5
18
+ },
19
+ {
20
+ "epoch": 0.056818181818181816,
21
+ "grad_norm": 1.082104889251594,
22
+ "learning_rate": 1.6666666666666667e-05,
23
+ "loss": 0.82,
24
+ "step": 10
25
+ },
26
+ {
27
+ "epoch": 0.08522727272727272,
28
+ "grad_norm": 0.6269246791788474,
29
+ "learning_rate": 2.5925925925925925e-05,
30
+ "loss": 0.7436,
31
+ "step": 15
32
+ },
33
+ {
34
+ "epoch": 0.11363636363636363,
35
+ "grad_norm": 0.416826795782172,
36
+ "learning_rate": 3.518518518518519e-05,
37
+ "loss": 0.7053,
38
+ "step": 20
39
+ },
40
+ {
41
+ "epoch": 0.14204545454545456,
42
+ "grad_norm": 0.33578935322162773,
43
+ "learning_rate": 4.4444444444444447e-05,
44
+ "loss": 0.6845,
45
+ "step": 25
46
+ },
47
+ {
48
+ "epoch": 0.17045454545454544,
49
+ "grad_norm": 0.2937906522794572,
50
+ "learning_rate": 4.99982305792533e-05,
51
+ "loss": 0.6605,
52
+ "step": 30
53
+ },
54
+ {
55
+ "epoch": 0.19886363636363635,
56
+ "grad_norm": 0.24747721249964458,
57
+ "learning_rate": 4.997832779179051e-05,
58
+ "loss": 0.637,
59
+ "step": 35
60
+ },
61
+ {
62
+ "epoch": 0.22727272727272727,
63
+ "grad_norm": 0.21048118380244446,
64
+ "learning_rate": 4.993633006948142e-05,
65
+ "loss": 0.638,
66
+ "step": 40
67
+ },
68
+ {
69
+ "epoch": 0.2556818181818182,
70
+ "grad_norm": 0.24332486977504197,
71
+ "learning_rate": 4.987227869373036e-05,
72
+ "loss": 0.6295,
73
+ "step": 45
74
+ },
75
+ {
76
+ "epoch": 0.2840909090909091,
77
+ "grad_norm": 0.3025474434568526,
78
+ "learning_rate": 4.9786236623445434e-05,
79
+ "loss": 0.6182,
80
+ "step": 50
81
+ },
82
+ {
83
+ "epoch": 0.3125,
84
+ "grad_norm": 0.3590729593937503,
85
+ "learning_rate": 4.967828843315348e-05,
86
+ "loss": 0.6123,
87
+ "step": 55
88
+ },
89
+ {
90
+ "epoch": 0.3409090909090909,
91
+ "grad_norm": 0.27055445475699047,
92
+ "learning_rate": 4.954854022986805e-05,
93
+ "loss": 0.6107,
94
+ "step": 60
95
+ },
96
+ {
97
+ "epoch": 0.3693181818181818,
98
+ "grad_norm": 0.5287101056579431,
99
+ "learning_rate": 4.939711954879211e-05,
100
+ "loss": 0.6076,
101
+ "step": 65
102
+ },
103
+ {
104
+ "epoch": 0.3977272727272727,
105
+ "grad_norm": 0.3411358973013703,
106
+ "learning_rate": 4.922417522795821e-05,
107
+ "loss": 0.607,
108
+ "step": 70
109
+ },
110
+ {
111
+ "epoch": 0.42613636363636365,
112
+ "grad_norm": 0.40275804442123037,
113
+ "learning_rate": 4.902987726192893e-05,
114
+ "loss": 0.5945,
115
+ "step": 75
116
+ },
117
+ {
118
+ "epoch": 0.45454545454545453,
119
+ "grad_norm": 0.45038950934752425,
120
+ "learning_rate": 4.881441663470182e-05,
121
+ "loss": 0.5993,
122
+ "step": 80
123
+ },
124
+ {
125
+ "epoch": 0.48295454545454547,
126
+ "grad_norm": 0.3017397368997767,
127
+ "learning_rate": 4.857800513198288e-05,
128
+ "loss": 0.591,
129
+ "step": 85
130
+ },
131
+ {
132
+ "epoch": 0.5113636363636364,
133
+ "grad_norm": 0.3047752307050808,
134
+ "learning_rate": 4.832087513301302e-05,
135
+ "loss": 0.5902,
136
+ "step": 90
137
+ },
138
+ {
139
+ "epoch": 0.5397727272727273,
140
+ "grad_norm": 0.35324983100208374,
141
+ "learning_rate": 4.804327938215235e-05,
142
+ "loss": 0.5882,
143
+ "step": 95
144
+ },
145
+ {
146
+ "epoch": 0.5681818181818182,
147
+ "grad_norm": 0.330403723283864,
148
+ "learning_rate": 4.7745490740446676e-05,
149
+ "loss": 0.5867,
150
+ "step": 100
151
+ },
152
+ {
153
+ "epoch": 0.5965909090909091,
154
+ "grad_norm": 0.31245134043332695,
155
+ "learning_rate": 4.742780191742036e-05,
156
+ "loss": 0.5851,
157
+ "step": 105
158
+ },
159
+ {
160
+ "epoch": 0.625,
161
+ "grad_norm": 0.33225122861646694,
162
+ "learning_rate": 4.7090525183359297e-05,
163
+ "loss": 0.5775,
164
+ "step": 110
165
+ },
166
+ {
167
+ "epoch": 0.6534090909090909,
168
+ "grad_norm": 0.3886059295138869,
169
+ "learning_rate": 4.6733992062366766e-05,
170
+ "loss": 0.5804,
171
+ "step": 115
172
+ },
173
+ {
174
+ "epoch": 0.6818181818181818,
175
+ "grad_norm": 0.3050360821401755,
176
+ "learning_rate": 4.635855300649382e-05,
177
+ "loss": 0.5772,
178
+ "step": 120
179
+ },
180
+ {
181
+ "epoch": 0.7102272727272727,
182
+ "grad_norm": 0.22763265674723182,
183
+ "learning_rate": 4.596457705126459e-05,
184
+ "loss": 0.5742,
185
+ "step": 125
186
+ },
187
+ {
188
+ "epoch": 0.7386363636363636,
189
+ "grad_norm": 0.25351069240901947,
190
+ "learning_rate": 4.555245145293509e-05,
191
+ "loss": 0.5777,
192
+ "step": 130
193
+ },
194
+ {
195
+ "epoch": 0.7670454545454546,
196
+ "grad_norm": 0.2912535971098546,
197
+ "learning_rate": 4.512258130784199e-05,
198
+ "loss": 0.5758,
199
+ "step": 135
200
+ },
201
+ {
202
+ "epoch": 0.7954545454545454,
203
+ "grad_norm": 0.2914231429766616,
204
+ "learning_rate": 4.4675389154215744e-05,
205
+ "loss": 0.5734,
206
+ "step": 140
207
+ },
208
+ {
209
+ "epoch": 0.8238636363636364,
210
+ "grad_norm": 0.33435962461349944,
211
+ "learning_rate": 4.4211314556849136e-05,
212
+ "loss": 0.5723,
213
+ "step": 145
214
+ },
215
+ {
216
+ "epoch": 0.8522727272727273,
217
+ "grad_norm": 0.36368454820155927,
218
+ "learning_rate": 4.373081367502981e-05,
219
+ "loss": 0.5715,
220
+ "step": 150
221
+ },
222
+ {
223
+ "epoch": 0.8806818181818182,
224
+ "grad_norm": 0.3129064534777125,
225
+ "learning_rate": 4.323435881416124e-05,
226
+ "loss": 0.5691,
227
+ "step": 155
228
+ },
229
+ {
230
+ "epoch": 0.9090909090909091,
231
+ "grad_norm": 0.42443011756875126,
232
+ "learning_rate": 4.272243796151305e-05,
233
+ "loss": 0.5665,
234
+ "step": 160
235
+ },
236
+ {
237
+ "epoch": 0.9375,
238
+ "grad_norm": 0.33009193150489613,
239
+ "learning_rate": 4.219555430655693e-05,
240
+ "loss": 0.5658,
241
+ "step": 165
242
+ },
243
+ {
244
+ "epoch": 0.9659090909090909,
245
+ "grad_norm": 0.3526624717761177,
246
+ "learning_rate": 4.165422574635959e-05,
247
+ "loss": 0.5678,
248
+ "step": 170
249
+ },
250
+ {
251
+ "epoch": 0.9943181818181818,
252
+ "grad_norm": 0.2673660036817638,
253
+ "learning_rate": 4.1098984376519e-05,
254
+ "loss": 0.5594,
255
+ "step": 175
256
+ },
257
+ {
258
+ "epoch": 1.0227272727272727,
259
+ "grad_norm": 0.34691003620364985,
260
+ "learning_rate": 4.053037596814432e-05,
261
+ "loss": 0.5476,
262
+ "step": 180
263
+ },
264
+ {
265
+ "epoch": 1.0511363636363635,
266
+ "grad_norm": 0.30758069223105644,
267
+ "learning_rate": 3.9948959431393454e-05,
268
+ "loss": 0.5413,
269
+ "step": 185
270
+ },
271
+ {
272
+ "epoch": 1.0795454545454546,
273
+ "grad_norm": 0.2927310952448428,
274
+ "learning_rate": 3.935530626609582e-05,
275
+ "loss": 0.5382,
276
+ "step": 190
277
+ },
278
+ {
279
+ "epoch": 1.1079545454545454,
280
+ "grad_norm": 0.2871242371453666,
281
+ "learning_rate": 3.875e-05,
282
+ "loss": 0.5391,
283
+ "step": 195
284
+ },
285
+ {
286
+ "epoch": 1.1363636363636362,
287
+ "grad_norm": 0.25692045940680436,
288
+ "learning_rate": 3.813363561519876e-05,
289
+ "loss": 0.5415,
290
+ "step": 200
291
+ },
292
+ {
293
+ "epoch": 1.1647727272727273,
294
+ "grad_norm": 0.24040159570359088,
295
+ "learning_rate": 3.750681896329504e-05,
296
+ "loss": 0.5426,
297
+ "step": 205
298
+ },
299
+ {
300
+ "epoch": 1.1931818181818181,
301
+ "grad_norm": 0.2888565204344371,
302
+ "learning_rate": 3.687016616988379e-05,
303
+ "loss": 0.5389,
304
+ "step": 210
305
+ },
306
+ {
307
+ "epoch": 1.2215909090909092,
308
+ "grad_norm": 0.23196268113152732,
309
+ "learning_rate": 3.6224303028935106e-05,
310
+ "loss": 0.537,
311
+ "step": 215
312
+ },
313
+ {
314
+ "epoch": 1.25,
315
+ "grad_norm": 0.2764992328867696,
316
+ "learning_rate": 3.556986438767389e-05,
317
+ "loss": 0.5412,
318
+ "step": 220
319
+ },
320
+ {
321
+ "epoch": 1.2784090909090908,
322
+ "grad_norm": 0.2846789099559735,
323
+ "learning_rate": 3.490749352256063e-05,
324
+ "loss": 0.5437,
325
+ "step": 225
326
+ },
327
+ {
328
+ "epoch": 1.3068181818181819,
329
+ "grad_norm": 0.23670429393756848,
330
+ "learning_rate": 3.4237841506986744e-05,
331
+ "loss": 0.5383,
332
+ "step": 230
333
+ },
334
+ {
335
+ "epoch": 1.3352272727272727,
336
+ "grad_norm": 0.3103095035403332,
337
+ "learning_rate": 3.356156657130596e-05,
338
+ "loss": 0.5349,
339
+ "step": 235
340
+ },
341
+ {
342
+ "epoch": 1.3636363636363638,
343
+ "grad_norm": 0.2134337589537462,
344
+ "learning_rate": 3.287933345583085e-05,
345
+ "loss": 0.5336,
346
+ "step": 240
347
+ },
348
+ {
349
+ "epoch": 1.3920454545454546,
350
+ "grad_norm": 0.24668367124517368,
351
+ "learning_rate": 3.219181275743034e-05,
352
+ "loss": 0.5311,
353
+ "step": 245
354
+ },
355
+ {
356
+ "epoch": 1.4204545454545454,
357
+ "grad_norm": 0.20533810672292555,
358
+ "learning_rate": 3.1499680270370706e-05,
359
+ "loss": 0.5362,
360
+ "step": 250
361
+ },
362
+ {
363
+ "epoch": 1.4488636363636362,
364
+ "grad_norm": 0.2521338151289615,
365
+ "learning_rate": 3.0803616322047654e-05,
366
+ "loss": 0.5377,
367
+ "step": 255
368
+ },
369
+ {
370
+ "epoch": 1.4772727272727273,
371
+ "grad_norm": 0.2172881728036589,
372
+ "learning_rate": 3.0104305104262698e-05,
373
+ "loss": 0.5301,
374
+ "step": 260
375
+ },
376
+ {
377
+ "epoch": 1.5056818181818183,
378
+ "grad_norm": 0.20056559882557096,
379
+ "learning_rate": 2.940243400070103e-05,
380
+ "loss": 0.532,
381
+ "step": 265
382
+ },
383
+ {
384
+ "epoch": 1.5340909090909092,
385
+ "grad_norm": 0.1775425002467347,
386
+ "learning_rate": 2.8698692911271897e-05,
387
+ "loss": 0.5303,
388
+ "step": 270
389
+ },
390
+ {
391
+ "epoch": 1.5625,
392
+ "grad_norm": 0.2109920372805477,
393
+ "learning_rate": 2.7993773573975728e-05,
394
+ "loss": 0.5317,
395
+ "step": 275
396
+ },
397
+ {
398
+ "epoch": 1.5909090909090908,
399
+ "grad_norm": 0.18902798919872466,
400
+ "learning_rate": 2.7288368884964475e-05,
401
+ "loss": 0.5325,
402
+ "step": 280
403
+ },
404
+ {
405
+ "epoch": 1.6193181818181817,
406
+ "grad_norm": 0.2186413096028553,
407
+ "learning_rate": 2.658317221746361e-05,
408
+ "loss": 0.5316,
409
+ "step": 285
410
+ },
411
+ {
412
+ "epoch": 1.6477272727272727,
413
+ "grad_norm": 0.21621244852102578,
414
+ "learning_rate": 2.5878876740225116e-05,
415
+ "loss": 0.5289,
416
+ "step": 290
417
+ },
418
+ {
419
+ "epoch": 1.6761363636363638,
420
+ "grad_norm": 0.19212370237001952,
421
+ "learning_rate": 2.5176174736181536e-05,
422
+ "loss": 0.5306,
423
+ "step": 295
424
+ },
425
+ {
426
+ "epoch": 1.7045454545454546,
427
+ "grad_norm": 0.183911038968379,
428
+ "learning_rate": 2.4475756921970653e-05,
429
+ "loss": 0.5353,
430
+ "step": 300
431
+ },
432
+ {
433
+ "epoch": 1.7329545454545454,
434
+ "grad_norm": 0.18937979590716766,
435
+ "learning_rate": 2.377831176899986e-05,
436
+ "loss": 0.5285,
437
+ "step": 305
438
+ },
439
+ {
440
+ "epoch": 1.7613636363636362,
441
+ "grad_norm": 0.24566289895185625,
442
+ "learning_rate": 2.3084524826717317e-05,
443
+ "loss": 0.5314,
444
+ "step": 310
445
+ },
446
+ {
447
+ "epoch": 1.7897727272727273,
448
+ "grad_norm": 0.1974602387281347,
449
+ "learning_rate": 2.23950780487554e-05,
450
+ "loss": 0.5271,
451
+ "step": 315
452
+ },
453
+ {
454
+ "epoch": 1.8181818181818183,
455
+ "grad_norm": 0.18954617375452634,
456
+ "learning_rate": 2.171064912260849e-05,
457
+ "loss": 0.5307,
458
+ "step": 320
459
+ },
460
+ {
461
+ "epoch": 1.8465909090909092,
462
+ "grad_norm": 0.19162555107387386,
463
+ "learning_rate": 2.103191080350419e-05,
464
+ "loss": 0.527,
465
+ "step": 325
466
+ },
467
+ {
468
+ "epoch": 1.875,
469
+ "grad_norm": 0.21263179284558476,
470
+ "learning_rate": 2.0359530253122738e-05,
471
+ "loss": 0.5291,
472
+ "step": 330
473
+ },
474
+ {
475
+ "epoch": 1.9034090909090908,
476
+ "grad_norm": 0.21095949546128553,
477
+ "learning_rate": 1.9694168383814492e-05,
478
+ "loss": 0.5246,
479
+ "step": 335
480
+ },
481
+ {
482
+ "epoch": 1.9318181818181817,
483
+ "grad_norm": 0.17656560903013435,
484
+ "learning_rate": 1.9036479208960127e-05,
485
+ "loss": 0.5207,
486
+ "step": 340
487
+ },
488
+ {
489
+ "epoch": 1.9602272727272727,
490
+ "grad_norm": 0.19556306442051233,
491
+ "learning_rate": 1.838710920011227e-05,
492
+ "loss": 0.5216,
493
+ "step": 345
494
+ },
495
+ {
496
+ "epoch": 1.9886363636363638,
497
+ "grad_norm": 0.19274133104006885,
498
+ "learning_rate": 1.7746696651550143e-05,
499
+ "loss": 0.5285,
500
+ "step": 350
501
+ },
502
+ {
503
+ "epoch": 2.0170454545454546,
504
+ "grad_norm": 0.21046167354718628,
505
+ "learning_rate": 1.7115871052872204e-05,
506
+ "loss": 0.5124,
507
+ "step": 355
508
+ },
509
+ {
510
+ "epoch": 2.0454545454545454,
511
+ "grad_norm": 0.18461635496510848,
512
+ "learning_rate": 1.6495252470243134e-05,
513
+ "loss": 0.5054,
514
+ "step": 360
515
+ },
516
+ {
517
+ "epoch": 2.0738636363636362,
518
+ "grad_norm": 0.18231246613717886,
519
+ "learning_rate": 1.5885450936903586e-05,
520
+ "loss": 0.5076,
521
+ "step": 365
522
+ },
523
+ {
524
+ "epoch": 2.102272727272727,
525
+ "grad_norm": 0.18193348469435636,
526
+ "learning_rate": 1.528706585354177e-05,
527
+ "loss": 0.506,
528
+ "step": 370
529
+ },
530
+ {
531
+ "epoch": 2.1306818181818183,
532
+ "grad_norm": 0.17073817731825536,
533
+ "learning_rate": 1.4700685399116238e-05,
534
+ "loss": 0.504,
535
+ "step": 375
536
+ },
537
+ {
538
+ "epoch": 2.159090909090909,
539
+ "grad_norm": 0.17027838857922944,
540
+ "learning_rate": 1.412688595270887e-05,
541
+ "loss": 0.5039,
542
+ "step": 380
543
+ },
544
+ {
545
+ "epoch": 2.1875,
546
+ "grad_norm": 0.18034007509851993,
547
+ "learning_rate": 1.3566231526976714e-05,
548
+ "loss": 0.5042,
549
+ "step": 385
550
+ },
551
+ {
552
+ "epoch": 2.215909090909091,
553
+ "grad_norm": 0.18256745032365382,
554
+ "learning_rate": 1.3019273213759081e-05,
555
+ "loss": 0.5073,
556
+ "step": 390
557
+ },
558
+ {
559
+ "epoch": 2.2443181818181817,
560
+ "grad_norm": 0.1818799435734757,
561
+ "learning_rate": 1.2486548642385274e-05,
562
+ "loss": 0.5041,
563
+ "step": 395
564
+ },
565
+ {
566
+ "epoch": 2.2727272727272725,
567
+ "grad_norm": 0.16719655038256215,
568
+ "learning_rate": 1.1968581451215065e-05,
569
+ "loss": 0.5052,
570
+ "step": 400
571
+ },
572
+ {
573
+ "epoch": 2.3011363636363638,
574
+ "grad_norm": 0.158314531917376,
575
+ "learning_rate": 1.1465880772931601e-05,
576
+ "loss": 0.5096,
577
+ "step": 405
578
+ },
579
+ {
580
+ "epoch": 2.3295454545454546,
581
+ "grad_norm": 0.16951382596927758,
582
+ "learning_rate": 1.0978940734092554e-05,
583
+ "loss": 0.5065,
584
+ "step": 410
585
+ },
586
+ {
587
+ "epoch": 2.3579545454545454,
588
+ "grad_norm": 0.14800468220648952,
589
+ "learning_rate": 1.050823996943144e-05,
590
+ "loss": 0.5068,
591
+ "step": 415
592
+ },
593
+ {
594
+ "epoch": 2.3863636363636362,
595
+ "grad_norm": 0.15562941027988406,
596
+ "learning_rate": 1.0054241151386492e-05,
597
+ "loss": 0.502,
598
+ "step": 420
599
+ },
600
+ {
601
+ "epoch": 2.4147727272727275,
602
+ "grad_norm": 0.15811584034908535,
603
+ "learning_rate": 9.617390535319684e-06,
604
+ "loss": 0.5057,
605
+ "step": 425
606
+ },
607
+ {
608
+ "epoch": 2.4431818181818183,
609
+ "grad_norm": 0.14436828305157706,
610
+ "learning_rate": 9.198117520872698e-06,
611
+ "loss": 0.502,
612
+ "step": 430
613
+ },
614
+ {
615
+ "epoch": 2.471590909090909,
616
+ "grad_norm": 0.15890187999735,
617
+ "learning_rate": 8.796834229891206e-06,
618
+ "loss": 0.5002,
619
+ "step": 435
620
+ },
621
+ {
622
+ "epoch": 2.5,
623
+ "grad_norm": 0.16251170123311687,
624
+ "learning_rate": 8.413935101332198e-06,
625
+ "loss": 0.5014,
626
+ "step": 440
627
+ },
628
+ {
629
+ "epoch": 2.528409090909091,
630
+ "grad_norm": 0.15588230293967764,
631
+ "learning_rate": 8.049796503552626e-06,
632
+ "loss": 0.5012,
633
+ "step": 445
634
+ },
635
+ {
636
+ "epoch": 2.5568181818181817,
637
+ "grad_norm": 0.1557285861591097,
638
+ "learning_rate": 7.704776364360454e-06,
639
+ "loss": 0.5072,
640
+ "step": 450
641
+ },
642
+ {
643
+ "epoch": 2.5852272727272725,
644
+ "grad_norm": 0.1502298284585181,
645
+ "learning_rate": 7.379213819191676e-06,
646
+ "loss": 0.507,
647
+ "step": 455
648
+ },
649
+ {
650
+ "epoch": 2.6136363636363638,
651
+ "grad_norm": 0.1558608079365694,
652
+ "learning_rate": 7.07342887775922e-06,
653
+ "loss": 0.5071,
654
+ "step": 460
655
+ },
656
+ {
657
+ "epoch": 2.6420454545454546,
658
+ "grad_norm": 0.15933885728264946,
659
+ "learning_rate": 6.787722109501379e-06,
660
+ "loss": 0.5074,
661
+ "step": 465
662
+ },
663
+ {
664
+ "epoch": 2.6704545454545454,
665
+ "grad_norm": 0.146461834299615,
666
+ "learning_rate": 6.522374348138882e-06,
667
+ "loss": 0.5038,
668
+ "step": 470
669
+ },
670
+ {
671
+ "epoch": 2.6988636363636362,
672
+ "grad_norm": 0.14024831946896077,
673
+ "learning_rate": 6.277646415631148e-06,
674
+ "loss": 0.5011,
675
+ "step": 475
676
+ },
677
+ {
678
+ "epoch": 2.7272727272727275,
679
+ "grad_norm": 0.1547607810382174,
680
+ "learning_rate": 6.053778865802907e-06,
681
+ "loss": 0.5076,
682
+ "step": 480
683
+ },
684
+ {
685
+ "epoch": 2.7556818181818183,
686
+ "grad_norm": 0.14445450539961152,
687
+ "learning_rate": 5.850991747893292e-06,
688
+ "loss": 0.5022,
689
+ "step": 485
690
+ },
691
+ {
692
+ "epoch": 2.784090909090909,
693
+ "grad_norm": 0.1502240146629453,
694
+ "learning_rate": 5.669484390259793e-06,
695
+ "loss": 0.5072,
696
+ "step": 490
697
+ },
698
+ {
699
+ "epoch": 2.8125,
700
+ "grad_norm": 0.14752200719540104,
701
+ "learning_rate": 5.509435204449666e-06,
702
+ "loss": 0.5025,
703
+ "step": 495
704
+ },
705
+ {
706
+ "epoch": 2.840909090909091,
707
+ "grad_norm": 0.13710947882432775,
708
+ "learning_rate": 5.37100150983139e-06,
709
+ "loss": 0.5014,
710
+ "step": 500
711
+ },
712
+ {
713
+ "epoch": 2.8693181818181817,
714
+ "grad_norm": 0.1434086525478373,
715
+ "learning_rate": 5.254319378958563e-06,
716
+ "loss": 0.5051,
717
+ "step": 505
718
+ },
719
+ {
720
+ "epoch": 2.8977272727272725,
721
+ "grad_norm": 0.13421579380890042,
722
+ "learning_rate": 5.159503503818199e-06,
723
+ "loss": 0.5024,
724
+ "step": 510
725
+ },
726
+ {
727
+ "epoch": 2.9261363636363638,
728
+ "grad_norm": 0.15773963719615386,
729
+ "learning_rate": 5.086647083094966e-06,
730
+ "loss": 0.5048,
731
+ "step": 515
732
+ },
733
+ {
734
+ "epoch": 2.9545454545454546,
735
+ "grad_norm": 0.1418412892614353,
736
+ "learning_rate": 5.03582173056209e-06,
737
+ "loss": 0.4993,
738
+ "step": 520
739
+ },
740
+ {
741
+ "epoch": 2.9829545454545454,
742
+ "grad_norm": 0.14410475241280002,
743
+ "learning_rate": 5.00707740468904e-06,
744
+ "loss": 0.4972,
745
+ "step": 525
746
+ },
747
+ {
748
+ "epoch": 3.0,
749
+ "step": 528,
750
+ "total_flos": 962656789856256.0,
751
+ "train_loss": 0.551490586363908,
752
+ "train_runtime": 11763.4703,
753
+ "train_samples_per_second": 5.731,
754
+ "train_steps_per_second": 0.045
755
+ }
756
+ ],
757
+ "logging_steps": 5,
758
+ "max_steps": 528,
759
+ "num_input_tokens_seen": 0,
760
+ "num_train_epochs": 3,
761
+ "save_steps": 100,
762
+ "stateful_callbacks": {
763
+ "TrainerControl": {
764
+ "args": {
765
+ "should_epoch_stop": false,
766
+ "should_evaluate": false,
767
+ "should_log": false,
768
+ "should_save": true,
769
+ "should_training_stop": true
770
+ },
771
+ "attributes": {}
772
+ }
773
+ },
774
+ "total_flos": 962656789856256.0,
775
+ "train_batch_size": 16,
776
+ "trial_name": null,
777
+ "trial_params": null
778
+ }