flyingbugs commited on
Commit
504d30d
·
verified ·
1 Parent(s): 8923f63

Model save

Browse files
README.md ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: Qwen/Qwen2.5-Math-7B-Instruct
3
+ library_name: transformers
4
+ model_name: Qwen2.5-Math-7B-s1k
5
+ tags:
6
+ - generated_from_trainer
7
+ - trl
8
+ - sft
9
+ licence: license
10
+ ---
11
+
12
+ # Model Card for Qwen2.5-Math-7B-s1k
13
+
14
+ This model is a fine-tuned version of [Qwen/Qwen2.5-Math-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-Math-7B-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-Math-7B-s1k", 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/x3m9pqj9)
31
+
32
+
33
+ This model was trained with SFT.
34
+
35
+ ### Framework versions
36
+
37
+ - TRL: 0.16.0.dev0
38
+ - Transformers: 4.49.0
39
+ - Pytorch: 2.5.1+cu121
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": 8.363833783064986e+16,
3
+ "train_loss": 1.3066247133981614,
4
+ "train_runtime": 6202.6685,
5
+ "train_samples": 1000,
6
+ "train_samples_per_second": 0.266,
7
+ "train_steps_per_second": 0.017
8
+ }
generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 151643,
3
+ "eos_token_id": [
4
+ 151645,
5
+ 151643
6
+ ],
7
+ "pad_token_id": 151643,
8
+ "transformers_version": "4.49.0"
9
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "total_flos": 8.363833783064986e+16,
3
+ "train_loss": 1.3066247133981614,
4
+ "train_runtime": 6202.6685,
5
+ "train_samples": 1000,
6
+ "train_samples_per_second": 0.266,
7
+ "train_steps_per_second": 0.017
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,777 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 5.0,
5
+ "eval_steps": 500,
6
+ "global_step": 105,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.047619047619047616,
13
+ "grad_norm": 46.73658663770524,
14
+ "learning_rate": 4.5454545454545455e-06,
15
+ "loss": 11.3797,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.09523809523809523,
20
+ "grad_norm": 44.523666214134316,
21
+ "learning_rate": 9.090909090909091e-06,
22
+ "loss": 11.5191,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.14285714285714285,
27
+ "grad_norm": 52.06639933238829,
28
+ "learning_rate": 1.3636363636363637e-05,
29
+ "loss": 10.8448,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.19047619047619047,
34
+ "grad_norm": 69.75134205890501,
35
+ "learning_rate": 1.8181818181818182e-05,
36
+ "loss": 9.5655,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.23809523809523808,
41
+ "grad_norm": 50.314368468329306,
42
+ "learning_rate": 2.272727272727273e-05,
43
+ "loss": 4.1848,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.2857142857142857,
48
+ "grad_norm": 7.357015429387816,
49
+ "learning_rate": 2.7272727272727273e-05,
50
+ "loss": 1.9862,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.3333333333333333,
55
+ "grad_norm": 7.401615020276034,
56
+ "learning_rate": 3.181818181818182e-05,
57
+ "loss": 1.9584,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.38095238095238093,
62
+ "grad_norm": 3.149164302427893,
63
+ "learning_rate": 3.6363636363636364e-05,
64
+ "loss": 1.5975,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.42857142857142855,
69
+ "grad_norm": 1.9054072928164427,
70
+ "learning_rate": 4.0909090909090915e-05,
71
+ "loss": 1.4777,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.47619047619047616,
76
+ "grad_norm": 2.09522071898415,
77
+ "learning_rate": 4.545454545454546e-05,
78
+ "loss": 1.4076,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.5238095238095238,
83
+ "grad_norm": 1.4886426113733002,
84
+ "learning_rate": 5e-05,
85
+ "loss": 1.3226,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 0.5714285714285714,
90
+ "grad_norm": 1.594161802836668,
91
+ "learning_rate": 4.946808510638298e-05,
92
+ "loss": 1.2825,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 0.6190476190476191,
97
+ "grad_norm": 1.5132463664090081,
98
+ "learning_rate": 4.893617021276596e-05,
99
+ "loss": 1.3821,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 0.6666666666666666,
104
+ "grad_norm": 1.4916977417841104,
105
+ "learning_rate": 4.840425531914894e-05,
106
+ "loss": 1.2297,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 0.7142857142857143,
111
+ "grad_norm": 1.07234396111405,
112
+ "learning_rate": 4.787234042553192e-05,
113
+ "loss": 1.2448,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 0.7619047619047619,
118
+ "grad_norm": 0.9261313011360983,
119
+ "learning_rate": 4.734042553191489e-05,
120
+ "loss": 1.2062,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 0.8095238095238095,
125
+ "grad_norm": 1.3648296835839082,
126
+ "learning_rate": 4.680851063829788e-05,
127
+ "loss": 1.2428,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 0.8571428571428571,
132
+ "grad_norm": 0.9239559683166827,
133
+ "learning_rate": 4.627659574468085e-05,
134
+ "loss": 1.2616,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 0.9047619047619048,
139
+ "grad_norm": 0.7209850490291895,
140
+ "learning_rate": 4.574468085106383e-05,
141
+ "loss": 1.0656,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 0.9523809523809523,
146
+ "grad_norm": 0.9769698482701535,
147
+ "learning_rate": 4.5212765957446815e-05,
148
+ "loss": 1.0884,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 1.0,
153
+ "grad_norm": 0.6106942050998588,
154
+ "learning_rate": 4.468085106382979e-05,
155
+ "loss": 1.0672,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 1.0476190476190477,
160
+ "grad_norm": 0.801040183534444,
161
+ "learning_rate": 4.414893617021277e-05,
162
+ "loss": 1.0478,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 1.0952380952380953,
167
+ "grad_norm": 0.7146659615588544,
168
+ "learning_rate": 4.3617021276595746e-05,
169
+ "loss": 1.091,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 1.1428571428571428,
174
+ "grad_norm": 0.5905709163842993,
175
+ "learning_rate": 4.3085106382978725e-05,
176
+ "loss": 1.1275,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 1.1904761904761905,
181
+ "grad_norm": 0.5826060194579513,
182
+ "learning_rate": 4.2553191489361704e-05,
183
+ "loss": 1.0079,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 1.2380952380952381,
188
+ "grad_norm": 0.6151603075503278,
189
+ "learning_rate": 4.2021276595744684e-05,
190
+ "loss": 1.0802,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 1.2857142857142856,
195
+ "grad_norm": 0.5264186877249164,
196
+ "learning_rate": 4.148936170212766e-05,
197
+ "loss": 1.0406,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 1.3333333333333333,
202
+ "grad_norm": 0.5461664948214178,
203
+ "learning_rate": 4.095744680851064e-05,
204
+ "loss": 1.0072,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 1.380952380952381,
209
+ "grad_norm": 0.4822129939360476,
210
+ "learning_rate": 4.0425531914893614e-05,
211
+ "loss": 1.0367,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 1.4285714285714286,
216
+ "grad_norm": 0.49848757883873196,
217
+ "learning_rate": 3.9893617021276594e-05,
218
+ "loss": 0.9754,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 1.4761904761904763,
223
+ "grad_norm": 0.5766625117987682,
224
+ "learning_rate": 3.936170212765958e-05,
225
+ "loss": 1.0631,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 1.5238095238095237,
230
+ "grad_norm": 0.4859811195017287,
231
+ "learning_rate": 3.882978723404255e-05,
232
+ "loss": 1.0003,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 1.5714285714285714,
237
+ "grad_norm": 0.5617547786833981,
238
+ "learning_rate": 3.829787234042553e-05,
239
+ "loss": 1.0312,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 1.619047619047619,
244
+ "grad_norm": 0.4760704205398701,
245
+ "learning_rate": 3.776595744680852e-05,
246
+ "loss": 0.9957,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 1.6666666666666665,
251
+ "grad_norm": 0.5525097760212518,
252
+ "learning_rate": 3.723404255319149e-05,
253
+ "loss": 1.0427,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 1.7142857142857144,
258
+ "grad_norm": 0.48722284467868165,
259
+ "learning_rate": 3.670212765957447e-05,
260
+ "loss": 0.9141,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 1.7619047619047619,
265
+ "grad_norm": 0.48286908931413486,
266
+ "learning_rate": 3.617021276595745e-05,
267
+ "loss": 0.9701,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 1.8095238095238095,
272
+ "grad_norm": 0.509607069236827,
273
+ "learning_rate": 3.563829787234043e-05,
274
+ "loss": 0.9234,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 1.8571428571428572,
279
+ "grad_norm": 0.48664273824178367,
280
+ "learning_rate": 3.5106382978723407e-05,
281
+ "loss": 1.0605,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 1.9047619047619047,
286
+ "grad_norm": 0.4951502177756041,
287
+ "learning_rate": 3.4574468085106386e-05,
288
+ "loss": 1.0418,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 1.9523809523809523,
293
+ "grad_norm": 0.5599844810380401,
294
+ "learning_rate": 3.4042553191489365e-05,
295
+ "loss": 1.0207,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 2.0,
300
+ "grad_norm": 0.47167217001219114,
301
+ "learning_rate": 3.3510638297872344e-05,
302
+ "loss": 0.8667,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 2.0476190476190474,
307
+ "grad_norm": 0.5766076030131434,
308
+ "learning_rate": 3.2978723404255317e-05,
309
+ "loss": 0.9305,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 2.0952380952380953,
314
+ "grad_norm": 0.5223243597728271,
315
+ "learning_rate": 3.2446808510638296e-05,
316
+ "loss": 0.8412,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 2.142857142857143,
321
+ "grad_norm": 0.46907426786479,
322
+ "learning_rate": 3.191489361702128e-05,
323
+ "loss": 0.8973,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 2.1904761904761907,
328
+ "grad_norm": 0.4701126779049834,
329
+ "learning_rate": 3.1382978723404254e-05,
330
+ "loss": 0.8509,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 2.238095238095238,
335
+ "grad_norm": 0.48418735511330596,
336
+ "learning_rate": 3.085106382978723e-05,
337
+ "loss": 0.8757,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 2.2857142857142856,
342
+ "grad_norm": 0.6556800657324943,
343
+ "learning_rate": 3.0319148936170216e-05,
344
+ "loss": 0.9124,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 2.3333333333333335,
349
+ "grad_norm": 0.6014700280019579,
350
+ "learning_rate": 2.9787234042553192e-05,
351
+ "loss": 0.8755,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 2.380952380952381,
356
+ "grad_norm": 0.5915165889418732,
357
+ "learning_rate": 2.925531914893617e-05,
358
+ "loss": 0.8925,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 2.4285714285714284,
363
+ "grad_norm": 0.6317703078383352,
364
+ "learning_rate": 2.8723404255319154e-05,
365
+ "loss": 0.8663,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 2.4761904761904763,
370
+ "grad_norm": 0.6429525648150871,
371
+ "learning_rate": 2.819148936170213e-05,
372
+ "loss": 0.8198,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 2.5238095238095237,
377
+ "grad_norm": 0.5187394614056622,
378
+ "learning_rate": 2.765957446808511e-05,
379
+ "loss": 0.7651,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 2.571428571428571,
384
+ "grad_norm": 0.4845593782104512,
385
+ "learning_rate": 2.7127659574468084e-05,
386
+ "loss": 0.8805,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 2.619047619047619,
391
+ "grad_norm": 0.47561465960794413,
392
+ "learning_rate": 2.6595744680851064e-05,
393
+ "loss": 0.9561,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 2.6666666666666665,
398
+ "grad_norm": 0.48208401639762166,
399
+ "learning_rate": 2.6063829787234046e-05,
400
+ "loss": 0.8638,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 2.7142857142857144,
405
+ "grad_norm": 0.5193206572882076,
406
+ "learning_rate": 2.5531914893617022e-05,
407
+ "loss": 0.8062,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 2.761904761904762,
412
+ "grad_norm": 0.5299016279110512,
413
+ "learning_rate": 2.5e-05,
414
+ "loss": 0.8236,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 2.8095238095238093,
419
+ "grad_norm": 0.47247578172991544,
420
+ "learning_rate": 2.446808510638298e-05,
421
+ "loss": 0.786,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 2.857142857142857,
426
+ "grad_norm": 0.4887209838922498,
427
+ "learning_rate": 2.393617021276596e-05,
428
+ "loss": 0.8275,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 2.9047619047619047,
433
+ "grad_norm": 0.47820835963456026,
434
+ "learning_rate": 2.340425531914894e-05,
435
+ "loss": 0.8581,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 2.9523809523809526,
440
+ "grad_norm": 0.46592054309550573,
441
+ "learning_rate": 2.2872340425531915e-05,
442
+ "loss": 0.8726,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 3.0,
447
+ "grad_norm": 0.5063226147267863,
448
+ "learning_rate": 2.2340425531914894e-05,
449
+ "loss": 0.8522,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 3.0476190476190474,
454
+ "grad_norm": 0.4349456733161818,
455
+ "learning_rate": 2.1808510638297873e-05,
456
+ "loss": 0.7678,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 3.0952380952380953,
461
+ "grad_norm": 0.43322041549114737,
462
+ "learning_rate": 2.1276595744680852e-05,
463
+ "loss": 0.7291,
464
+ "step": 65
465
+ },
466
+ {
467
+ "epoch": 3.142857142857143,
468
+ "grad_norm": 0.3891658154700991,
469
+ "learning_rate": 2.074468085106383e-05,
470
+ "loss": 0.7098,
471
+ "step": 66
472
+ },
473
+ {
474
+ "epoch": 3.1904761904761907,
475
+ "grad_norm": 0.48090626334768466,
476
+ "learning_rate": 2.0212765957446807e-05,
477
+ "loss": 0.7075,
478
+ "step": 67
479
+ },
480
+ {
481
+ "epoch": 3.238095238095238,
482
+ "grad_norm": 0.4192109123095055,
483
+ "learning_rate": 1.968085106382979e-05,
484
+ "loss": 0.7135,
485
+ "step": 68
486
+ },
487
+ {
488
+ "epoch": 3.2857142857142856,
489
+ "grad_norm": 0.4538447830174013,
490
+ "learning_rate": 1.9148936170212766e-05,
491
+ "loss": 0.7293,
492
+ "step": 69
493
+ },
494
+ {
495
+ "epoch": 3.3333333333333335,
496
+ "grad_norm": 0.400428247237971,
497
+ "learning_rate": 1.8617021276595745e-05,
498
+ "loss": 0.7182,
499
+ "step": 70
500
+ },
501
+ {
502
+ "epoch": 3.380952380952381,
503
+ "grad_norm": 0.41259082028208505,
504
+ "learning_rate": 1.8085106382978724e-05,
505
+ "loss": 0.7111,
506
+ "step": 71
507
+ },
508
+ {
509
+ "epoch": 3.4285714285714284,
510
+ "grad_norm": 0.42210992879504494,
511
+ "learning_rate": 1.7553191489361703e-05,
512
+ "loss": 0.7358,
513
+ "step": 72
514
+ },
515
+ {
516
+ "epoch": 3.4761904761904763,
517
+ "grad_norm": 0.4805171232293903,
518
+ "learning_rate": 1.7021276595744682e-05,
519
+ "loss": 0.7693,
520
+ "step": 73
521
+ },
522
+ {
523
+ "epoch": 3.5238095238095237,
524
+ "grad_norm": 0.4290739751199803,
525
+ "learning_rate": 1.6489361702127658e-05,
526
+ "loss": 0.7538,
527
+ "step": 74
528
+ },
529
+ {
530
+ "epoch": 3.571428571428571,
531
+ "grad_norm": 0.46386967645771476,
532
+ "learning_rate": 1.595744680851064e-05,
533
+ "loss": 0.7483,
534
+ "step": 75
535
+ },
536
+ {
537
+ "epoch": 3.619047619047619,
538
+ "grad_norm": 0.4600000815160992,
539
+ "learning_rate": 1.5425531914893617e-05,
540
+ "loss": 0.6961,
541
+ "step": 76
542
+ },
543
+ {
544
+ "epoch": 3.6666666666666665,
545
+ "grad_norm": 0.48967690124437346,
546
+ "learning_rate": 1.4893617021276596e-05,
547
+ "loss": 0.7759,
548
+ "step": 77
549
+ },
550
+ {
551
+ "epoch": 3.7142857142857144,
552
+ "grad_norm": 0.43799987175594934,
553
+ "learning_rate": 1.4361702127659577e-05,
554
+ "loss": 0.6973,
555
+ "step": 78
556
+ },
557
+ {
558
+ "epoch": 3.761904761904762,
559
+ "grad_norm": 0.4170501746303817,
560
+ "learning_rate": 1.3829787234042554e-05,
561
+ "loss": 0.7713,
562
+ "step": 79
563
+ },
564
+ {
565
+ "epoch": 3.8095238095238093,
566
+ "grad_norm": 0.44313691170789904,
567
+ "learning_rate": 1.3297872340425532e-05,
568
+ "loss": 0.7287,
569
+ "step": 80
570
+ },
571
+ {
572
+ "epoch": 3.857142857142857,
573
+ "grad_norm": 0.3626173621626196,
574
+ "learning_rate": 1.2765957446808511e-05,
575
+ "loss": 0.7152,
576
+ "step": 81
577
+ },
578
+ {
579
+ "epoch": 3.9047619047619047,
580
+ "grad_norm": 0.40980441661962524,
581
+ "learning_rate": 1.223404255319149e-05,
582
+ "loss": 0.7388,
583
+ "step": 82
584
+ },
585
+ {
586
+ "epoch": 3.9523809523809526,
587
+ "grad_norm": 0.3476970829860624,
588
+ "learning_rate": 1.170212765957447e-05,
589
+ "loss": 0.6835,
590
+ "step": 83
591
+ },
592
+ {
593
+ "epoch": 4.0,
594
+ "grad_norm": 0.34534853622038764,
595
+ "learning_rate": 1.1170212765957447e-05,
596
+ "loss": 0.6936,
597
+ "step": 84
598
+ },
599
+ {
600
+ "epoch": 4.0476190476190474,
601
+ "grad_norm": 0.34717149811141956,
602
+ "learning_rate": 1.0638297872340426e-05,
603
+ "loss": 0.6412,
604
+ "step": 85
605
+ },
606
+ {
607
+ "epoch": 4.095238095238095,
608
+ "grad_norm": 0.39121639106624856,
609
+ "learning_rate": 1.0106382978723404e-05,
610
+ "loss": 0.6447,
611
+ "step": 86
612
+ },
613
+ {
614
+ "epoch": 4.142857142857143,
615
+ "grad_norm": 0.3853359648131134,
616
+ "learning_rate": 9.574468085106383e-06,
617
+ "loss": 0.67,
618
+ "step": 87
619
+ },
620
+ {
621
+ "epoch": 4.190476190476191,
622
+ "grad_norm": 0.333219732679469,
623
+ "learning_rate": 9.042553191489362e-06,
624
+ "loss": 0.6951,
625
+ "step": 88
626
+ },
627
+ {
628
+ "epoch": 4.238095238095238,
629
+ "grad_norm": 0.4113440938373367,
630
+ "learning_rate": 8.510638297872341e-06,
631
+ "loss": 0.6427,
632
+ "step": 89
633
+ },
634
+ {
635
+ "epoch": 4.285714285714286,
636
+ "grad_norm": 0.4292570295550588,
637
+ "learning_rate": 7.97872340425532e-06,
638
+ "loss": 0.6563,
639
+ "step": 90
640
+ },
641
+ {
642
+ "epoch": 4.333333333333333,
643
+ "grad_norm": 0.3325469730092871,
644
+ "learning_rate": 7.446808510638298e-06,
645
+ "loss": 0.6215,
646
+ "step": 91
647
+ },
648
+ {
649
+ "epoch": 4.380952380952381,
650
+ "grad_norm": 0.37390136106392785,
651
+ "learning_rate": 6.914893617021277e-06,
652
+ "loss": 0.6194,
653
+ "step": 92
654
+ },
655
+ {
656
+ "epoch": 4.428571428571429,
657
+ "grad_norm": 0.4143640454410716,
658
+ "learning_rate": 6.3829787234042555e-06,
659
+ "loss": 0.6207,
660
+ "step": 93
661
+ },
662
+ {
663
+ "epoch": 4.476190476190476,
664
+ "grad_norm": 0.3648244167487935,
665
+ "learning_rate": 5.851063829787235e-06,
666
+ "loss": 0.5998,
667
+ "step": 94
668
+ },
669
+ {
670
+ "epoch": 4.523809523809524,
671
+ "grad_norm": 0.3512446333779434,
672
+ "learning_rate": 5.319148936170213e-06,
673
+ "loss": 0.6351,
674
+ "step": 95
675
+ },
676
+ {
677
+ "epoch": 4.571428571428571,
678
+ "grad_norm": 0.339436273470992,
679
+ "learning_rate": 4.787234042553191e-06,
680
+ "loss": 0.599,
681
+ "step": 96
682
+ },
683
+ {
684
+ "epoch": 4.619047619047619,
685
+ "grad_norm": 0.4440038854080703,
686
+ "learning_rate": 4.255319148936171e-06,
687
+ "loss": 0.584,
688
+ "step": 97
689
+ },
690
+ {
691
+ "epoch": 4.666666666666667,
692
+ "grad_norm": 0.36581621694617583,
693
+ "learning_rate": 3.723404255319149e-06,
694
+ "loss": 0.6203,
695
+ "step": 98
696
+ },
697
+ {
698
+ "epoch": 4.714285714285714,
699
+ "grad_norm": 0.3509331585886843,
700
+ "learning_rate": 3.1914893617021277e-06,
701
+ "loss": 0.5981,
702
+ "step": 99
703
+ },
704
+ {
705
+ "epoch": 4.761904761904762,
706
+ "grad_norm": 0.3403976301106126,
707
+ "learning_rate": 2.6595744680851065e-06,
708
+ "loss": 0.6415,
709
+ "step": 100
710
+ },
711
+ {
712
+ "epoch": 4.809523809523809,
713
+ "grad_norm": 0.30854856192103863,
714
+ "learning_rate": 2.1276595744680853e-06,
715
+ "loss": 0.6054,
716
+ "step": 101
717
+ },
718
+ {
719
+ "epoch": 4.857142857142857,
720
+ "grad_norm": 0.2984920819758894,
721
+ "learning_rate": 1.5957446808510639e-06,
722
+ "loss": 0.6145,
723
+ "step": 102
724
+ },
725
+ {
726
+ "epoch": 4.904761904761905,
727
+ "grad_norm": 0.30989076720361786,
728
+ "learning_rate": 1.0638297872340427e-06,
729
+ "loss": 0.6317,
730
+ "step": 103
731
+ },
732
+ {
733
+ "epoch": 4.9523809523809526,
734
+ "grad_norm": 0.3438826262811614,
735
+ "learning_rate": 5.319148936170213e-07,
736
+ "loss": 0.6193,
737
+ "step": 104
738
+ },
739
+ {
740
+ "epoch": 5.0,
741
+ "grad_norm": 0.32644346890582215,
742
+ "learning_rate": 0.0,
743
+ "loss": 0.6277,
744
+ "step": 105
745
+ },
746
+ {
747
+ "epoch": 5.0,
748
+ "step": 105,
749
+ "total_flos": 8.363833783064986e+16,
750
+ "train_loss": 1.3066247133981614,
751
+ "train_runtime": 6202.6685,
752
+ "train_samples_per_second": 0.266,
753
+ "train_steps_per_second": 0.017
754
+ }
755
+ ],
756
+ "logging_steps": 1,
757
+ "max_steps": 105,
758
+ "num_input_tokens_seen": 0,
759
+ "num_train_epochs": 5,
760
+ "save_steps": 500,
761
+ "stateful_callbacks": {
762
+ "TrainerControl": {
763
+ "args": {
764
+ "should_epoch_stop": false,
765
+ "should_evaluate": false,
766
+ "should_log": false,
767
+ "should_save": true,
768
+ "should_training_stop": true
769
+ },
770
+ "attributes": {}
771
+ }
772
+ },
773
+ "total_flos": 8.363833783064986e+16,
774
+ "train_batch_size": 1,
775
+ "trial_name": null,
776
+ "trial_params": null
777
+ }