gordonguanghui commited on
Commit
139efe6
·
verified ·
1 Parent(s): 74219cd

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
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
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="gordonguanghui/Qwen2.5-1.5B-Open-R1-Distill", 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/wangguanghui0607/huggingface/runs/rva174tq)
31
+
32
+
33
+ This model was trained with SFT.
34
+
35
+ ### Framework versions
36
+
37
+ - TRL: 0.15.0
38
+ - Transformers: 4.49.0
39
+ - Pytorch: 2.5.1
40
+ - Datasets: 3.3.2
41
+ - Tokenizers: 0.21.0
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": 488621249396736.0,
3
+ "train_loss": 0.5612850879777723,
4
+ "train_runtime": 11235.1008,
5
+ "train_samples": 93733,
6
+ "train_samples_per_second": 3.053,
7
+ "train_steps_per_second": 0.048
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": "4.49.0"
14
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "total_flos": 488621249396736.0,
3
+ "train_loss": 0.5612850879777723,
4
+ "train_runtime": 11235.1008,
5
+ "train_samples": 93733,
6
+ "train_samples_per_second": 3.053,
7
+ "train_steps_per_second": 0.048
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,791 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 536,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.009328358208955223,
13
+ "grad_norm": 1.4534666584342113,
14
+ "learning_rate": 9.259259259259259e-06,
15
+ "loss": 0.8491,
16
+ "step": 5
17
+ },
18
+ {
19
+ "epoch": 0.018656716417910446,
20
+ "grad_norm": 1.1711916406560432,
21
+ "learning_rate": 1.8518518518518518e-05,
22
+ "loss": 0.8005,
23
+ "step": 10
24
+ },
25
+ {
26
+ "epoch": 0.027985074626865673,
27
+ "grad_norm": 0.7520916620233523,
28
+ "learning_rate": 2.777777777777778e-05,
29
+ "loss": 0.7313,
30
+ "step": 15
31
+ },
32
+ {
33
+ "epoch": 0.03731343283582089,
34
+ "grad_norm": 0.4675969780487494,
35
+ "learning_rate": 3.7037037037037037e-05,
36
+ "loss": 0.6802,
37
+ "step": 20
38
+ },
39
+ {
40
+ "epoch": 0.04664179104477612,
41
+ "grad_norm": 0.47274993852668085,
42
+ "learning_rate": 4.62962962962963e-05,
43
+ "loss": 0.6577,
44
+ "step": 25
45
+ },
46
+ {
47
+ "epoch": 0.055970149253731345,
48
+ "grad_norm": 0.4212378058530289,
49
+ "learning_rate": 4.999614302517356e-05,
50
+ "loss": 0.658,
51
+ "step": 30
52
+ },
53
+ {
54
+ "epoch": 0.06529850746268656,
55
+ "grad_norm": 0.3974247959215564,
56
+ "learning_rate": 4.997257741198456e-05,
57
+ "loss": 0.6419,
58
+ "step": 35
59
+ },
60
+ {
61
+ "epoch": 0.07462686567164178,
62
+ "grad_norm": 0.3207977209065435,
63
+ "learning_rate": 4.992761136351291e-05,
64
+ "loss": 0.627,
65
+ "step": 40
66
+ },
67
+ {
68
+ "epoch": 0.08395522388059702,
69
+ "grad_norm": 0.34035380085701855,
70
+ "learning_rate": 4.986128770052603e-05,
71
+ "loss": 0.6084,
72
+ "step": 45
73
+ },
74
+ {
75
+ "epoch": 0.09328358208955224,
76
+ "grad_norm": 0.30354586362391395,
77
+ "learning_rate": 4.9773669582457364e-05,
78
+ "loss": 0.6101,
79
+ "step": 50
80
+ },
81
+ {
82
+ "epoch": 0.10261194029850747,
83
+ "grad_norm": 0.3866792254018339,
84
+ "learning_rate": 4.966484044726024e-05,
85
+ "loss": 0.6138,
86
+ "step": 55
87
+ },
88
+ {
89
+ "epoch": 0.11194029850746269,
90
+ "grad_norm": 0.37949830558516523,
91
+ "learning_rate": 4.953490393195063e-05,
92
+ "loss": 0.6006,
93
+ "step": 60
94
+ },
95
+ {
96
+ "epoch": 0.12126865671641791,
97
+ "grad_norm": 0.41179707666908344,
98
+ "learning_rate": 4.938398377391461e-05,
99
+ "loss": 0.5975,
100
+ "step": 65
101
+ },
102
+ {
103
+ "epoch": 0.13059701492537312,
104
+ "grad_norm": 0.5822016464409348,
105
+ "learning_rate": 4.921222369307427e-05,
106
+ "loss": 0.5941,
107
+ "step": 70
108
+ },
109
+ {
110
+ "epoch": 0.13992537313432835,
111
+ "grad_norm": 0.4786878847425219,
112
+ "learning_rate": 4.901978725502454e-05,
113
+ "loss": 0.5897,
114
+ "step": 75
115
+ },
116
+ {
117
+ "epoch": 0.14925373134328357,
118
+ "grad_norm": 0.4033723216228573,
119
+ "learning_rate": 4.880685771527114e-05,
120
+ "loss": 0.5896,
121
+ "step": 80
122
+ },
123
+ {
124
+ "epoch": 0.15858208955223882,
125
+ "grad_norm": 0.45477274659957395,
126
+ "learning_rate": 4.8573637844718e-05,
127
+ "loss": 0.5946,
128
+ "step": 85
129
+ },
130
+ {
131
+ "epoch": 0.16791044776119404,
132
+ "grad_norm": 0.4591987379848635,
133
+ "learning_rate": 4.83203497365703e-05,
134
+ "loss": 0.5913,
135
+ "step": 90
136
+ },
137
+ {
138
+ "epoch": 0.17723880597014927,
139
+ "grad_norm": 0.512818276549736,
140
+ "learning_rate": 4.8047234594837143e-05,
141
+ "loss": 0.5783,
142
+ "step": 95
143
+ },
144
+ {
145
+ "epoch": 0.1865671641791045,
146
+ "grad_norm": 0.4685138256637361,
147
+ "learning_rate": 4.775455250463507e-05,
148
+ "loss": 0.5751,
149
+ "step": 100
150
+ },
151
+ {
152
+ "epoch": 0.1958955223880597,
153
+ "grad_norm": 0.418117783574535,
154
+ "learning_rate": 4.744258218451135e-05,
155
+ "loss": 0.5849,
156
+ "step": 105
157
+ },
158
+ {
159
+ "epoch": 0.20522388059701493,
160
+ "grad_norm": 0.29569410470796165,
161
+ "learning_rate": 4.71116207210228e-05,
162
+ "loss": 0.5761,
163
+ "step": 110
164
+ },
165
+ {
166
+ "epoch": 0.21455223880597016,
167
+ "grad_norm": 0.36588086743641784,
168
+ "learning_rate": 4.676198328582288e-05,
169
+ "loss": 0.5713,
170
+ "step": 115
171
+ },
172
+ {
173
+ "epoch": 0.22388059701492538,
174
+ "grad_norm": 0.3341648966276976,
175
+ "learning_rate": 4.6394002835526535e-05,
176
+ "loss": 0.5822,
177
+ "step": 120
178
+ },
179
+ {
180
+ "epoch": 0.2332089552238806,
181
+ "grad_norm": 0.41521121469362104,
182
+ "learning_rate": 4.6008029794638596e-05,
183
+ "loss": 0.5668,
184
+ "step": 125
185
+ },
186
+ {
187
+ "epoch": 0.24253731343283583,
188
+ "grad_norm": 0.3961894491982948,
189
+ "learning_rate": 4.560443172184763e-05,
190
+ "loss": 0.5694,
191
+ "step": 130
192
+ },
193
+ {
194
+ "epoch": 0.251865671641791,
195
+ "grad_norm": 0.3162166728580678,
196
+ "learning_rate": 4.5183592960003104e-05,
197
+ "loss": 0.562,
198
+ "step": 135
199
+ },
200
+ {
201
+ "epoch": 0.26119402985074625,
202
+ "grad_norm": 0.30927428386203404,
203
+ "learning_rate": 4.4745914270109055e-05,
204
+ "loss": 0.5673,
205
+ "step": 140
206
+ },
207
+ {
208
+ "epoch": 0.27052238805970147,
209
+ "grad_norm": 0.3536678799668726,
210
+ "learning_rate": 4.429181244968301e-05,
211
+ "loss": 0.5665,
212
+ "step": 145
213
+ },
214
+ {
215
+ "epoch": 0.2798507462686567,
216
+ "grad_norm": 0.33334109534004264,
217
+ "learning_rate": 4.38217199358434e-05,
218
+ "loss": 0.5605,
219
+ "step": 150
220
+ },
221
+ {
222
+ "epoch": 0.2891791044776119,
223
+ "grad_norm": 0.322653801766586,
224
+ "learning_rate": 4.3336084393503545e-05,
225
+ "loss": 0.5569,
226
+ "step": 155
227
+ },
228
+ {
229
+ "epoch": 0.29850746268656714,
230
+ "grad_norm": 0.3821614756667243,
231
+ "learning_rate": 4.283536828906436e-05,
232
+ "loss": 0.5693,
233
+ "step": 160
234
+ },
235
+ {
236
+ "epoch": 0.30783582089552236,
237
+ "grad_norm": 0.3833424325515942,
238
+ "learning_rate": 4.2320048450011684e-05,
239
+ "loss": 0.5628,
240
+ "step": 165
241
+ },
242
+ {
243
+ "epoch": 0.31716417910447764,
244
+ "grad_norm": 0.28204498731498767,
245
+ "learning_rate": 4.179061561083777e-05,
246
+ "loss": 0.5549,
247
+ "step": 170
248
+ },
249
+ {
250
+ "epoch": 0.32649253731343286,
251
+ "grad_norm": 0.4006876246661576,
252
+ "learning_rate": 4.124757394571914e-05,
253
+ "loss": 0.5613,
254
+ "step": 175
255
+ },
256
+ {
257
+ "epoch": 0.3358208955223881,
258
+ "grad_norm": 0.2991827926881174,
259
+ "learning_rate": 4.069144058839605e-05,
260
+ "loss": 0.5623,
261
+ "step": 180
262
+ },
263
+ {
264
+ "epoch": 0.3451492537313433,
265
+ "grad_norm": 0.36867658282010896,
266
+ "learning_rate": 4.012274513971061e-05,
267
+ "loss": 0.5665,
268
+ "step": 185
269
+ },
270
+ {
271
+ "epoch": 0.35447761194029853,
272
+ "grad_norm": 0.34656580201404646,
273
+ "learning_rate": 3.954202916327264e-05,
274
+ "loss": 0.5543,
275
+ "step": 190
276
+ },
277
+ {
278
+ "epoch": 0.36380597014925375,
279
+ "grad_norm": 0.32173574580213904,
280
+ "learning_rate": 3.894984566973346e-05,
281
+ "loss": 0.5545,
282
+ "step": 195
283
+ },
284
+ {
285
+ "epoch": 0.373134328358209,
286
+ "grad_norm": 0.29550755846319327,
287
+ "learning_rate": 3.834675859015876e-05,
288
+ "loss": 0.5584,
289
+ "step": 200
290
+ },
291
+ {
292
+ "epoch": 0.3824626865671642,
293
+ "grad_norm": 0.2981084523878612,
294
+ "learning_rate": 3.77333422390021e-05,
295
+ "loss": 0.5492,
296
+ "step": 205
297
+ },
298
+ {
299
+ "epoch": 0.3917910447761194,
300
+ "grad_norm": 0.3082421191846,
301
+ "learning_rate": 3.711018076719034e-05,
302
+ "loss": 0.5564,
303
+ "step": 210
304
+ },
305
+ {
306
+ "epoch": 0.40111940298507465,
307
+ "grad_norm": 0.3079273697879876,
308
+ "learning_rate": 3.647786760584194e-05,
309
+ "loss": 0.5607,
310
+ "step": 215
311
+ },
312
+ {
313
+ "epoch": 0.41044776119402987,
314
+ "grad_norm": 0.33213977610506984,
315
+ "learning_rate": 3.583700490114776e-05,
316
+ "loss": 0.5588,
317
+ "step": 220
318
+ },
319
+ {
320
+ "epoch": 0.4197761194029851,
321
+ "grad_norm": 0.3454312169959124,
322
+ "learning_rate": 3.518820294095267e-05,
323
+ "loss": 0.5548,
324
+ "step": 225
325
+ },
326
+ {
327
+ "epoch": 0.4291044776119403,
328
+ "grad_norm": 0.3111048651445072,
329
+ "learning_rate": 3.453207957358377e-05,
330
+ "loss": 0.5467,
331
+ "step": 230
332
+ },
333
+ {
334
+ "epoch": 0.43843283582089554,
335
+ "grad_norm": 0.3122159165573577,
336
+ "learning_rate": 3.386925961947906e-05,
337
+ "loss": 0.5478,
338
+ "step": 235
339
+ },
340
+ {
341
+ "epoch": 0.44776119402985076,
342
+ "grad_norm": 0.3815095659501264,
343
+ "learning_rate": 3.320037427617639e-05,
344
+ "loss": 0.5553,
345
+ "step": 240
346
+ },
347
+ {
348
+ "epoch": 0.457089552238806,
349
+ "grad_norm": 0.29471688390864853,
350
+ "learning_rate": 3.252606051722972e-05,
351
+ "loss": 0.5538,
352
+ "step": 245
353
+ },
354
+ {
355
+ "epoch": 0.4664179104477612,
356
+ "grad_norm": 0.3449459267022439,
357
+ "learning_rate": 3.1846960485624886e-05,
358
+ "loss": 0.5429,
359
+ "step": 250
360
+ },
361
+ {
362
+ "epoch": 0.47574626865671643,
363
+ "grad_norm": 0.3845030239986481,
364
+ "learning_rate": 3.1163720882272516e-05,
365
+ "loss": 0.5507,
366
+ "step": 255
367
+ },
368
+ {
369
+ "epoch": 0.48507462686567165,
370
+ "grad_norm": 0.3473121194428009,
371
+ "learning_rate": 3.047699235016056e-05,
372
+ "loss": 0.543,
373
+ "step": 260
374
+ },
375
+ {
376
+ "epoch": 0.4944029850746269,
377
+ "grad_norm": 0.3183251881853174,
378
+ "learning_rate": 2.9787428854752736e-05,
379
+ "loss": 0.5368,
380
+ "step": 265
381
+ },
382
+ {
383
+ "epoch": 0.503731343283582,
384
+ "grad_norm": 0.28586873082292064,
385
+ "learning_rate": 2.9095687061223058e-05,
386
+ "loss": 0.5516,
387
+ "step": 270
388
+ },
389
+ {
390
+ "epoch": 0.5130597014925373,
391
+ "grad_norm": 0.34819444255911386,
392
+ "learning_rate": 2.8402425709119435e-05,
393
+ "loss": 0.5506,
394
+ "step": 275
395
+ },
396
+ {
397
+ "epoch": 0.5223880597014925,
398
+ "grad_norm": 0.2704108425263057,
399
+ "learning_rate": 2.7708304985051868e-05,
400
+ "loss": 0.5477,
401
+ "step": 280
402
+ },
403
+ {
404
+ "epoch": 0.5317164179104478,
405
+ "grad_norm": 0.28669112746143444,
406
+ "learning_rate": 2.7013985894002623e-05,
407
+ "loss": 0.5356,
408
+ "step": 285
409
+ },
410
+ {
411
+ "epoch": 0.5410447761194029,
412
+ "grad_norm": 0.2920252669556323,
413
+ "learning_rate": 2.6320129629857093e-05,
414
+ "loss": 0.5483,
415
+ "step": 290
416
+ },
417
+ {
418
+ "epoch": 0.5503731343283582,
419
+ "grad_norm": 0.2706020130328637,
420
+ "learning_rate": 2.56273969457547e-05,
421
+ "loss": 0.5373,
422
+ "step": 295
423
+ },
424
+ {
425
+ "epoch": 0.5597014925373134,
426
+ "grad_norm": 0.26211562115767434,
427
+ "learning_rate": 2.4936447524859625e-05,
428
+ "loss": 0.5416,
429
+ "step": 300
430
+ },
431
+ {
432
+ "epoch": 0.5690298507462687,
433
+ "grad_norm": 0.32486475734880277,
434
+ "learning_rate": 2.4247939352150386e-05,
435
+ "loss": 0.5367,
436
+ "step": 305
437
+ },
438
+ {
439
+ "epoch": 0.5783582089552238,
440
+ "grad_norm": 0.26365125864733874,
441
+ "learning_rate": 2.3562528087826573e-05,
442
+ "loss": 0.5429,
443
+ "step": 310
444
+ },
445
+ {
446
+ "epoch": 0.5876865671641791,
447
+ "grad_norm": 0.2513654842729793,
448
+ "learning_rate": 2.2880866442929544e-05,
449
+ "loss": 0.5398,
450
+ "step": 315
451
+ },
452
+ {
453
+ "epoch": 0.5970149253731343,
454
+ "grad_norm": 0.2429254461380733,
455
+ "learning_rate": 2.2203603557771447e-05,
456
+ "loss": 0.5359,
457
+ "step": 320
458
+ },
459
+ {
460
+ "epoch": 0.6063432835820896,
461
+ "grad_norm": 0.26495785362725305,
462
+ "learning_rate": 2.153138438376473e-05,
463
+ "loss": 0.5342,
464
+ "step": 325
465
+ },
466
+ {
467
+ "epoch": 0.6156716417910447,
468
+ "grad_norm": 0.24698377526769538,
469
+ "learning_rate": 2.0864849069240645e-05,
470
+ "loss": 0.5389,
471
+ "step": 330
472
+ },
473
+ {
474
+ "epoch": 0.625,
475
+ "grad_norm": 0.2716935501857811,
476
+ "learning_rate": 2.0204632349841667e-05,
477
+ "loss": 0.5357,
478
+ "step": 335
479
+ },
480
+ {
481
+ "epoch": 0.6343283582089553,
482
+ "grad_norm": 0.23513260099346342,
483
+ "learning_rate": 1.9551362944068462e-05,
484
+ "loss": 0.538,
485
+ "step": 340
486
+ },
487
+ {
488
+ "epoch": 0.6436567164179104,
489
+ "grad_norm": 0.26133157599374934,
490
+ "learning_rate": 1.890566295455678e-05,
491
+ "loss": 0.5313,
492
+ "step": 345
493
+ },
494
+ {
495
+ "epoch": 0.6529850746268657,
496
+ "grad_norm": 0.2449995496020128,
497
+ "learning_rate": 1.8268147275654707e-05,
498
+ "loss": 0.5411,
499
+ "step": 350
500
+ },
501
+ {
502
+ "epoch": 0.6623134328358209,
503
+ "grad_norm": 0.2247533847427069,
504
+ "learning_rate": 1.7639423007864252e-05,
505
+ "loss": 0.5281,
506
+ "step": 355
507
+ },
508
+ {
509
+ "epoch": 0.6716417910447762,
510
+ "grad_norm": 0.2260024812533735,
511
+ "learning_rate": 1.702008887970491e-05,
512
+ "loss": 0.5289,
513
+ "step": 360
514
+ },
515
+ {
516
+ "epoch": 0.6809701492537313,
517
+ "grad_norm": 0.23618301620256973,
518
+ "learning_rate": 1.6410734677549872e-05,
519
+ "loss": 0.5322,
520
+ "step": 365
521
+ },
522
+ {
523
+ "epoch": 0.6902985074626866,
524
+ "grad_norm": 0.24235987409675933,
525
+ "learning_rate": 1.58119406839777e-05,
526
+ "loss": 0.5311,
527
+ "step": 370
528
+ },
529
+ {
530
+ "epoch": 0.6996268656716418,
531
+ "grad_norm": 0.2559397079642031,
532
+ "learning_rate": 1.5224277125174388e-05,
533
+ "loss": 0.534,
534
+ "step": 375
535
+ },
536
+ {
537
+ "epoch": 0.7089552238805971,
538
+ "grad_norm": 0.2276944155439983,
539
+ "learning_rate": 1.464830362791204e-05,
540
+ "loss": 0.5363,
541
+ "step": 380
542
+ },
543
+ {
544
+ "epoch": 0.7182835820895522,
545
+ "grad_norm": 0.2199325902339066,
546
+ "learning_rate": 1.4084568686621314e-05,
547
+ "loss": 0.5386,
548
+ "step": 385
549
+ },
550
+ {
551
+ "epoch": 0.7276119402985075,
552
+ "grad_norm": 0.2188363369888737,
553
+ "learning_rate": 1.3533609141065008e-05,
554
+ "loss": 0.5368,
555
+ "step": 390
556
+ },
557
+ {
558
+ "epoch": 0.7369402985074627,
559
+ "grad_norm": 0.244089061405243,
560
+ "learning_rate": 1.299594966511038e-05,
561
+ "loss": 0.5341,
562
+ "step": 395
563
+ },
564
+ {
565
+ "epoch": 0.746268656716418,
566
+ "grad_norm": 0.22680536791217729,
567
+ "learning_rate": 1.2472102267086904e-05,
568
+ "loss": 0.5297,
569
+ "step": 400
570
+ },
571
+ {
572
+ "epoch": 0.7555970149253731,
573
+ "grad_norm": 0.2167560534054679,
574
+ "learning_rate": 1.1962565802205255e-05,
575
+ "loss": 0.5364,
576
+ "step": 405
577
+ },
578
+ {
579
+ "epoch": 0.7649253731343284,
580
+ "grad_norm": 0.2069416144688831,
581
+ "learning_rate": 1.1467825497501954e-05,
582
+ "loss": 0.5228,
583
+ "step": 410
584
+ },
585
+ {
586
+ "epoch": 0.7742537313432836,
587
+ "grad_norm": 0.2302676785561096,
588
+ "learning_rate": 1.0988352489762006e-05,
589
+ "loss": 0.5387,
590
+ "step": 415
591
+ },
592
+ {
593
+ "epoch": 0.7835820895522388,
594
+ "grad_norm": 0.2170000089714264,
595
+ "learning_rate": 1.052460337685951e-05,
596
+ "loss": 0.5301,
597
+ "step": 420
598
+ },
599
+ {
600
+ "epoch": 0.792910447761194,
601
+ "grad_norm": 0.21330280289188058,
602
+ "learning_rate": 1.0077019782943584e-05,
603
+ "loss": 0.5285,
604
+ "step": 425
605
+ },
606
+ {
607
+ "epoch": 0.8022388059701493,
608
+ "grad_norm": 0.22789671376553575,
609
+ "learning_rate": 9.646027937883622e-06,
610
+ "loss": 0.5293,
611
+ "step": 430
612
+ },
613
+ {
614
+ "epoch": 0.8115671641791045,
615
+ "grad_norm": 0.2044388742976582,
616
+ "learning_rate": 9.232038271374377e-06,
617
+ "loss": 0.5312,
618
+ "step": 435
619
+ },
620
+ {
621
+ "epoch": 0.8208955223880597,
622
+ "grad_norm": 0.2121630276342737,
623
+ "learning_rate": 8.835445022087426e-06,
624
+ "loss": 0.5259,
625
+ "step": 440
626
+ },
627
+ {
628
+ "epoch": 0.8302238805970149,
629
+ "grad_norm": 0.21958440427518724,
630
+ "learning_rate": 8.456625862241193e-06,
631
+ "loss": 0.5361,
632
+ "step": 445
633
+ },
634
+ {
635
+ "epoch": 0.8395522388059702,
636
+ "grad_norm": 0.1993684141564352,
637
+ "learning_rate": 8.095941537947057e-06,
638
+ "loss": 0.5331,
639
+ "step": 450
640
+ },
641
+ {
642
+ "epoch": 0.8488805970149254,
643
+ "grad_norm": 0.22924886203423017,
644
+ "learning_rate": 7.753735525674059e-06,
645
+ "loss": 0.5258,
646
+ "step": 455
647
+ },
648
+ {
649
+ "epoch": 0.8582089552238806,
650
+ "grad_norm": 0.20514748374762057,
651
+ "learning_rate": 7.430333705159286e-06,
652
+ "loss": 0.5361,
653
+ "step": 460
654
+ },
655
+ {
656
+ "epoch": 0.8675373134328358,
657
+ "grad_norm": 0.19989921284599924,
658
+ "learning_rate": 7.126044049075548e-06,
659
+ "loss": 0.541,
660
+ "step": 465
661
+ },
662
+ {
663
+ "epoch": 0.8768656716417911,
664
+ "grad_norm": 0.19586385268397843,
665
+ "learning_rate": 6.8411563297516995e-06,
666
+ "loss": 0.5211,
667
+ "step": 470
668
+ },
669
+ {
670
+ "epoch": 0.8861940298507462,
671
+ "grad_norm": 0.1864657226226325,
672
+ "learning_rate": 6.575941843225068e-06,
673
+ "loss": 0.5249,
674
+ "step": 475
675
+ },
676
+ {
677
+ "epoch": 0.8955223880597015,
678
+ "grad_norm": 0.2190651926378328,
679
+ "learning_rate": 6.330653150888617e-06,
680
+ "loss": 0.5294,
681
+ "step": 480
682
+ },
683
+ {
684
+ "epoch": 0.9048507462686567,
685
+ "grad_norm": 0.2138242740322654,
686
+ "learning_rate": 6.105523838979022e-06,
687
+ "loss": 0.5375,
688
+ "step": 485
689
+ },
690
+ {
691
+ "epoch": 0.914179104477612,
692
+ "grad_norm": 0.19520032583669467,
693
+ "learning_rate": 5.900768296134551e-06,
694
+ "loss": 0.524,
695
+ "step": 490
696
+ },
697
+ {
698
+ "epoch": 0.9235074626865671,
699
+ "grad_norm": 0.1918945141121369,
700
+ "learning_rate": 5.7165815092346825e-06,
701
+ "loss": 0.5263,
702
+ "step": 495
703
+ },
704
+ {
705
+ "epoch": 0.9328358208955224,
706
+ "grad_norm": 0.18952383838163894,
707
+ "learning_rate": 5.553138877715833e-06,
708
+ "loss": 0.5281,
709
+ "step": 500
710
+ },
711
+ {
712
+ "epoch": 0.9421641791044776,
713
+ "grad_norm": 0.19849132534942884,
714
+ "learning_rate": 5.410596046540051e-06,
715
+ "loss": 0.5232,
716
+ "step": 505
717
+ },
718
+ {
719
+ "epoch": 0.9514925373134329,
720
+ "grad_norm": 0.21507884974780903,
721
+ "learning_rate": 5.28908875797568e-06,
722
+ "loss": 0.5237,
723
+ "step": 510
724
+ },
725
+ {
726
+ "epoch": 0.960820895522388,
727
+ "grad_norm": 0.19952690630617984,
728
+ "learning_rate": 5.1887327223312296e-06,
729
+ "loss": 0.5248,
730
+ "step": 515
731
+ },
732
+ {
733
+ "epoch": 0.9701492537313433,
734
+ "grad_norm": 0.2050907612918201,
735
+ "learning_rate": 5.109623507765466e-06,
736
+ "loss": 0.5291,
737
+ "step": 520
738
+ },
739
+ {
740
+ "epoch": 0.9794776119402985,
741
+ "grad_norm": 0.20534363644652603,
742
+ "learning_rate": 5.051836449278715e-06,
743
+ "loss": 0.5313,
744
+ "step": 525
745
+ },
746
+ {
747
+ "epoch": 0.9888059701492538,
748
+ "grad_norm": 0.20717020550482595,
749
+ "learning_rate": 5.015426576972003e-06,
750
+ "loss": 0.5301,
751
+ "step": 530
752
+ },
753
+ {
754
+ "epoch": 0.9981343283582089,
755
+ "grad_norm": 0.190446884081656,
756
+ "learning_rate": 5.000428563642382e-06,
757
+ "loss": 0.5302,
758
+ "step": 535
759
+ },
760
+ {
761
+ "epoch": 1.0,
762
+ "step": 536,
763
+ "total_flos": 488621249396736.0,
764
+ "train_loss": 0.5612850879777723,
765
+ "train_runtime": 11235.1008,
766
+ "train_samples_per_second": 3.053,
767
+ "train_steps_per_second": 0.048
768
+ }
769
+ ],
770
+ "logging_steps": 5,
771
+ "max_steps": 536,
772
+ "num_input_tokens_seen": 0,
773
+ "num_train_epochs": 1,
774
+ "save_steps": 100,
775
+ "stateful_callbacks": {
776
+ "TrainerControl": {
777
+ "args": {
778
+ "should_epoch_stop": false,
779
+ "should_evaluate": false,
780
+ "should_log": false,
781
+ "should_save": true,
782
+ "should_training_stop": true
783
+ },
784
+ "attributes": {}
785
+ }
786
+ },
787
+ "total_flos": 488621249396736.0,
788
+ "train_batch_size": 8,
789
+ "trial_name": null,
790
+ "trial_params": null
791
+ }