chinna6 commited on
Commit
7a3b8aa
·
verified ·
1 Parent(s): 5f8d9b2

End of training

Browse files
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:73ce040e47bc2b66aed9044bd62426c52ee86454b3332da4702bbcc78d486395
3
  size 35237104
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ec94e591006394e5861342585e4aaf983b29a686e71b397bcc3aa5e4b210244d
3
  size 35237104
all_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "total_flos": 0.0,
3
- "train_loss": 2.2126714176451402e-07,
4
- "train_runtime": 2214.7036,
5
  "train_samples": 31,
6
- "train_samples_per_second": 0.433,
7
- "train_steps_per_second": 0.027
8
  }
 
1
  {
2
  "total_flos": 0.0,
3
+ "train_loss": 1.1757511778398565e-06,
4
+ "train_runtime": 2092.798,
5
  "train_samples": 31,
6
+ "train_samples_per_second": 0.459,
7
+ "train_steps_per_second": 0.029
8
  }
train_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "total_flos": 0.0,
3
- "train_loss": 2.2126714176451402e-07,
4
- "train_runtime": 2214.7036,
5
  "train_samples": 31,
6
- "train_samples_per_second": 0.433,
7
- "train_steps_per_second": 0.027
8
  }
 
1
  {
2
  "total_flos": 0.0,
3
+ "train_loss": 1.1757511778398565e-06,
4
+ "train_runtime": 2092.798,
5
  "train_samples": 31,
6
+ "train_samples_per_second": 0.459,
7
+ "train_steps_per_second": 0.029
8
  }
trainer_state.json CHANGED
@@ -10,583 +10,583 @@
10
  "is_world_process_zero": true,
11
  "log_history": [
12
  {
13
- "completion_length": 468.09375,
14
  "epoch": 0.25806451612903225,
15
  "grad_norm": NaN,
16
- "kl": 0.00012564308224227716,
17
  "learning_rate": 2.5e-07,
18
  "loss": 0.0,
19
- "reward": 0.1961599849164486,
20
- "reward_std": 0.37325981876347214,
21
  "rewards/concensus_correctness_reward_func": 0.0,
22
  "rewards/consensus_reward_func": 0.0,
23
  "rewards/cumulative_reward_2": 0.0,
24
  "rewards/final_correctness_reward_func": 0.0,
25
- "rewards/question_recreation_reward_func": 0.135066231014207,
26
  "rewards/soft_format_reward_func": 0.0,
27
  "rewards/strict_format_reward_func": 0.0,
28
- "rewards/xmlcount_reward_func": 0.06109375273808837,
29
  "step": 2
30
  },
31
  {
32
- "completion_length": 394.59375,
33
  "epoch": 0.5161290322580645,
34
- "grad_norm": 2.8862016201019287,
35
- "kl": 3.452185706009914e-05,
36
  "learning_rate": 5e-07,
37
  "loss": 0.0,
38
- "reward": 0.11793552408926189,
39
- "reward_std": 0.16145133040845394,
40
  "rewards/concensus_correctness_reward_func": 0.0,
41
  "rewards/consensus_reward_func": 0.0,
42
  "rewards/cumulative_reward_2": 0.0,
43
  "rewards/final_correctness_reward_func": 0.0,
44
- "rewards/question_recreation_reward_func": 0.08818553062155843,
45
  "rewards/soft_format_reward_func": 0.0,
46
  "rewards/strict_format_reward_func": 0.0,
47
- "rewards/xmlcount_reward_func": 0.029749999288469553,
48
  "step": 4
49
  },
50
  {
51
- "completion_length": 363.40625,
52
  "epoch": 0.7741935483870968,
53
- "grad_norm": 2.1103923320770264,
54
- "kl": 2.3821680770197418e-05,
55
- "learning_rate": 4.996333534627809e-07,
56
  "loss": 0.0,
57
- "reward": 0.09433670341968536,
58
- "reward_std": 0.26609469903633,
59
  "rewards/concensus_correctness_reward_func": 0.0,
60
  "rewards/consensus_reward_func": 0.0,
61
  "rewards/cumulative_reward_2": 0.0,
62
  "rewards/final_correctness_reward_func": 0.0,
63
- "rewards/question_recreation_reward_func": 0.1436179461888969,
64
  "rewards/soft_format_reward_func": 0.0,
65
  "rewards/strict_format_reward_func": 0.0,
66
- "rewards/xmlcount_reward_func": -0.049281246960163116,
67
  "step": 6
68
  },
69
  {
70
- "completion_length": 374.57142857142856,
71
  "epoch": 1.0,
72
- "grad_norm": 1.734000325202942,
73
- "kl": 1.5630724841945005e-05,
74
  "learning_rate": 4.967066306353815e-07,
75
  "loss": 0.0,
76
- "reward": 0.2236826547554561,
77
- "reward_std": 0.17228200073753083,
78
  "rewards/concensus_correctness_reward_func": 0.0,
79
  "rewards/consensus_reward_func": 0.0,
80
  "rewards/cumulative_reward_2": 0.0,
81
  "rewards/final_correctness_reward_func": 0.0,
82
- "rewards/question_recreation_reward_func": 0.13896837245140756,
83
  "rewards/soft_format_reward_func": 0.0,
84
  "rewards/strict_format_reward_func": 0.0,
85
- "rewards/xmlcount_reward_func": 0.08471428602933884,
86
  "step": 8
87
  },
88
  {
89
- "completion_length": 418.46875,
90
  "epoch": 1.2580645161290323,
91
- "grad_norm": 1.0722944736480713,
92
- "kl": 3.292944302302203e-05,
93
- "learning_rate": 4.908874981298057e-07,
94
  "loss": 0.0,
95
- "reward": 0.1162781547755003,
96
- "reward_std": 0.6191137628629804,
97
  "rewards/concensus_correctness_reward_func": 0.0,
98
  "rewards/consensus_reward_func": 0.0,
99
  "rewards/cumulative_reward_2": 0.0,
100
  "rewards/final_correctness_reward_func": 0.0,
101
- "rewards/question_recreation_reward_func": 0.2167781526222825,
102
  "rewards/soft_format_reward_func": 0.0,
103
  "rewards/strict_format_reward_func": 0.0,
104
- "rewards/xmlcount_reward_func": -0.1004999983124435,
105
  "step": 10
106
  },
107
  {
108
- "completion_length": 442.96875,
109
  "epoch": 1.5161290322580645,
110
- "grad_norm": 16.9965763092041,
111
- "kl": 7.811630212017917e-05,
112
  "learning_rate": 4.869132927957006e-07,
113
  "loss": 0.0,
114
- "reward": 0.1594895573798567,
115
- "reward_std": 0.14911856758408248,
116
  "rewards/concensus_correctness_reward_func": 0.0,
117
  "rewards/consensus_reward_func": 0.0,
118
  "rewards/cumulative_reward_2": 0.0,
119
  "rewards/final_correctness_reward_func": 0.0,
120
- "rewards/question_recreation_reward_func": 0.10480205074418336,
121
  "rewards/soft_format_reward_func": 0.0,
122
  "rewards/strict_format_reward_func": 0.0,
123
- "rewards/xmlcount_reward_func": 0.0546875,
124
  "step": 12
125
  },
126
  {
127
- "completion_length": 383.15625,
128
  "epoch": 1.7741935483870968,
129
- "grad_norm": 3.1423959732055664,
130
- "kl": 7.733380130048317e-05,
131
  "learning_rate": 4.768938549177392e-07,
132
  "loss": 0.0,
133
- "reward": 0.17886506160721183,
134
- "reward_std": 0.34254320128820837,
135
  "rewards/concensus_correctness_reward_func": 0.0,
136
  "rewards/consensus_reward_func": 0.0,
137
  "rewards/cumulative_reward_2": 0.0,
138
  "rewards/final_correctness_reward_func": 0.0,
139
- "rewards/question_recreation_reward_func": 0.1495525580830872,
140
  "rewards/soft_format_reward_func": 0.0,
141
  "rewards/strict_format_reward_func": 0.0,
142
- "rewards/xmlcount_reward_func": 0.02931250142864883,
143
  "step": 14
144
  },
145
  {
146
- "completion_length": 468.2857142857143,
147
  "epoch": 2.0,
148
- "grad_norm": 1.208578109741211,
149
- "kl": 1.6717019986702197e-05,
150
- "learning_rate": 4.642142940418973e-07,
151
- "loss": -0.0,
152
- "reward": 0.30199881323746275,
153
- "reward_std": 0.3142883365175554,
154
  "rewards/concensus_correctness_reward_func": 0.0,
155
  "rewards/consensus_reward_func": 0.0,
156
  "rewards/cumulative_reward_2": 0.0,
157
  "rewards/final_correctness_reward_func": 0.0,
158
- "rewards/question_recreation_reward_func": 0.16321310134870665,
159
  "rewards/soft_format_reward_func": 0.0,
160
  "rewards/strict_format_reward_func": 0.0,
161
- "rewards/xmlcount_reward_func": 0.13878571508186205,
162
  "step": 16
163
  },
164
  {
165
- "completion_length": 338.84375,
166
  "epoch": 2.258064516129032,
167
- "grad_norm": 1.581721544265747,
168
- "kl": 8.628947284705646e-05,
169
  "learning_rate": 4.5692224953922265e-07,
170
  "loss": 0.0,
171
- "reward": 0.2258664125110954,
172
- "reward_std": 0.18049531092401594,
173
  "rewards/concensus_correctness_reward_func": 0.0,
174
  "rewards/consensus_reward_func": 0.0,
175
  "rewards/cumulative_reward_2": 0.0,
176
  "rewards/final_correctness_reward_func": 0.0,
177
- "rewards/question_recreation_reward_func": 0.08808515965938568,
178
  "rewards/soft_format_reward_func": 0.0,
179
  "rewards/strict_format_reward_func": 0.0,
180
- "rewards/xmlcount_reward_func": 0.13778125029057264,
181
  "step": 18
182
  },
183
  {
184
- "completion_length": 454.125,
185
  "epoch": 2.5161290322580645,
186
- "grad_norm": 2.681896209716797,
187
- "kl": 3.0871672720422794e-05,
188
  "learning_rate": 4.405405137819091e-07,
189
  "loss": 0.0,
190
- "reward": -0.02695425506681204,
191
- "reward_std": 0.7458917489275336,
192
  "rewards/concensus_correctness_reward_func": 0.0,
193
  "rewards/consensus_reward_func": 0.0,
194
  "rewards/cumulative_reward_2": 0.0,
195
  "rewards/final_correctness_reward_func": 0.0,
196
- "rewards/question_recreation_reward_func": 0.1082957349717617,
197
  "rewards/soft_format_reward_func": 0.0,
198
  "rewards/strict_format_reward_func": 0.0,
199
- "rewards/xmlcount_reward_func": -0.1352499956265092,
200
  "step": 20
201
  },
202
  {
203
- "completion_length": 363.40625,
204
  "epoch": 2.774193548387097,
205
- "grad_norm": 3.3509066104888916,
206
- "kl": 0.00016107357282635348,
207
  "learning_rate": 4.219248647133558e-07,
208
  "loss": 0.0,
209
- "reward": 0.29551874240860343,
210
- "reward_std": 0.26216102205216885,
211
  "rewards/concensus_correctness_reward_func": 0.0,
212
  "rewards/consensus_reward_func": 0.0,
213
  "rewards/cumulative_reward_2": 0.0,
214
  "rewards/final_correctness_reward_func": 0.0,
215
- "rewards/question_recreation_reward_func": 0.16439374536275864,
216
  "rewards/soft_format_reward_func": 0.0,
217
  "rewards/strict_format_reward_func": 0.0,
218
- "rewards/xmlcount_reward_func": 0.13112499960698187,
219
  "step": 22
220
  },
221
  {
222
- "completion_length": 440.82142857142856,
223
  "epoch": 3.0,
224
- "grad_norm": 6.132716178894043,
225
- "kl": 2.4475816254770117e-05,
226
  "learning_rate": 4.012935537984413e-07,
227
  "loss": 0.0,
228
- "reward": 0.17054232954978943,
229
- "reward_std": 0.742070411997182,
230
  "rewards/concensus_correctness_reward_func": 0.0,
231
  "rewards/consensus_reward_func": 0.0,
232
  "rewards/cumulative_reward_2": 0.0,
233
  "rewards/final_correctness_reward_func": 0.0,
234
- "rewards/question_recreation_reward_func": 0.2551851751548903,
235
  "rewards/soft_format_reward_func": 0.0,
236
  "rewards/strict_format_reward_func": 0.0,
237
- "rewards/xmlcount_reward_func": -0.08464285651487964,
238
  "step": 24
239
  },
240
  {
241
- "completion_length": 376.3125,
242
  "epoch": 3.258064516129032,
243
- "grad_norm": 18.50641632080078,
244
- "kl": 0.0002098049449159589,
245
  "learning_rate": 3.788884642942554e-07,
246
  "loss": 0.0,
247
- "reward": 0.36438849940896034,
248
- "reward_std": 0.43193568650167435,
249
  "rewards/concensus_correctness_reward_func": 0.0,
250
  "rewards/consensus_reward_func": 0.0,
251
  "rewards/cumulative_reward_2": 0.0,
252
  "rewards/final_correctness_reward_func": 0.0,
253
- "rewards/question_recreation_reward_func": 0.2423259963397868,
254
  "rewards/soft_format_reward_func": 0.0,
255
  "rewards/strict_format_reward_func": 0.0,
256
- "rewards/xmlcount_reward_func": 0.12206250196322799,
257
  "step": 26
258
  },
259
  {
260
- "completion_length": 470.875,
261
  "epoch": 3.5161290322580645,
262
- "grad_norm": 30.481353759765625,
263
- "kl": 0.00012239812804182293,
264
  "learning_rate": 3.5497227539006614e-07,
265
  "loss": 0.0,
266
- "reward": 0.2246184330433607,
267
- "reward_std": 0.25769937818404287,
268
- "rewards/concensus_correctness_reward_func": 0.0,
269
  "rewards/consensus_reward_func": 0.0,
270
  "rewards/cumulative_reward_2": 0.0,
271
  "rewards/final_correctness_reward_func": 0.0,
272
- "rewards/question_recreation_reward_func": 0.09374343021772802,
273
  "rewards/soft_format_reward_func": 0.0,
274
  "rewards/strict_format_reward_func": 0.0,
275
- "rewards/xmlcount_reward_func": 0.1308750007301569,
276
  "step": 28
277
  },
278
  {
279
- "completion_length": 489.3125,
280
  "epoch": 3.774193548387097,
281
- "grad_norm": 1.8558696508407593,
282
- "kl": 7.136619524317211e-05,
283
  "learning_rate": 3.2982538253399494e-07,
284
  "loss": 0.0,
285
- "reward": 0.3110757367685437,
286
- "reward_std": 0.3201977536082268,
287
  "rewards/concensus_correctness_reward_func": 0.0,
288
  "rewards/consensus_reward_func": 0.0,
289
  "rewards/cumulative_reward_2": 0.0,
290
  "rewards/final_correctness_reward_func": 0.0,
291
- "rewards/question_recreation_reward_func": 0.21076323743909597,
292
  "rewards/soft_format_reward_func": 0.0,
293
  "rewards/strict_format_reward_func": 0.0,
294
- "rewards/xmlcount_reward_func": 0.10031249932944775,
295
  "step": 30
296
  },
297
  {
298
- "completion_length": 375.7857142857143,
299
  "epoch": 4.0,
300
- "grad_norm": 6.766973495483398,
301
- "kl": 6.908967420063814e-05,
302
  "learning_rate": 3.0374261005275605e-07,
303
  "loss": 0.0,
304
- "reward": 0.006137807454381671,
305
- "reward_std": 0.27292134180398925,
306
  "rewards/concensus_correctness_reward_func": 0.0,
307
  "rewards/consensus_reward_func": 0.0,
308
  "rewards/cumulative_reward_2": 0.0,
309
  "rewards/final_correctness_reward_func": 0.0,
310
- "rewards/question_recreation_reward_func": 0.08913780535970416,
311
  "rewards/soft_format_reward_func": 0.0,
312
  "rewards/strict_format_reward_func": 0.0,
313
- "rewards/xmlcount_reward_func": -0.08299999790532249,
314
  "step": 32
315
  },
316
  {
317
- "completion_length": 415.15625,
318
  "epoch": 4.258064516129032,
319
- "grad_norm": 1.7305625677108765,
320
- "kl": 2.464489864451025e-05,
321
  "learning_rate": 2.770297546059854e-07,
322
  "loss": 0.0,
323
- "reward": 0.3909291820600629,
324
- "reward_std": 0.48921840731054544,
325
  "rewards/concensus_correctness_reward_func": 0.0,
326
  "rewards/consensus_reward_func": 0.0,
327
  "rewards/cumulative_reward_2": 0.0,
328
  "rewards/final_correctness_reward_func": 0.0,
329
- "rewards/question_recreation_reward_func": 0.22102294443175197,
330
  "rewards/soft_format_reward_func": 0.0,
331
  "rewards/strict_format_reward_func": 0.0,
332
- "rewards/xmlcount_reward_func": 0.16990624740719795,
333
  "step": 34
334
  },
335
  {
336
- "completion_length": 403.96875,
337
  "epoch": 4.516129032258064,
338
- "grad_norm": 2.19710636138916,
339
- "kl": 4.924387462779123e-05,
340
  "learning_rate": 2.5e-07,
341
  "loss": 0.0,
342
- "reward": 0.13032389804720879,
343
- "reward_std": 0.2979316939599812,
344
  "rewards/concensus_correctness_reward_func": 0.0,
345
  "rewards/consensus_reward_func": 0.0,
346
  "rewards/cumulative_reward_2": 0.0,
347
  "rewards/final_correctness_reward_func": 0.0,
348
- "rewards/question_recreation_reward_func": 0.11326138750882819,
349
  "rewards/soft_format_reward_func": 0.0,
350
  "rewards/strict_format_reward_func": 0.0,
351
- "rewards/xmlcount_reward_func": 0.01706250198185444,
352
  "step": 36
353
  },
354
  {
355
- "completion_length": 518.46875,
356
  "epoch": 4.774193548387097,
357
- "grad_norm": 22.284698486328125,
358
- "kl": 0.000312702951873689,
359
  "learning_rate": 2.229702453940146e-07,
360
  "loss": 0.0,
361
- "reward": 0.16360057145357132,
362
- "reward_std": 0.40589657705277205,
363
  "rewards/concensus_correctness_reward_func": 0.0,
364
  "rewards/consensus_reward_func": 0.0,
365
  "rewards/cumulative_reward_2": 0.0,
366
  "rewards/final_correctness_reward_func": 0.0,
367
- "rewards/question_recreation_reward_func": 0.11656931973993778,
368
  "rewards/soft_format_reward_func": 0.0,
369
  "rewards/strict_format_reward_func": 0.0,
370
- "rewards/xmlcount_reward_func": 0.04703124985098839,
371
  "step": 38
372
  },
373
  {
374
- "completion_length": 217.14285714285714,
375
  "epoch": 5.0,
376
- "grad_norm": 15.186713218688965,
377
- "kl": 0.0013402944018058147,
378
  "learning_rate": 1.9625738994724398e-07,
379
  "loss": 0.0,
380
- "reward": 0.14894756089363778,
381
- "reward_std": 0.21149546493377006,
382
  "rewards/concensus_correctness_reward_func": 0.0,
383
  "rewards/consensus_reward_func": 0.0,
384
  "rewards/cumulative_reward_2": 0.0,
385
  "rewards/final_correctness_reward_func": 0.0,
386
- "rewards/question_recreation_reward_func": 0.15655470479811942,
387
  "rewards/soft_format_reward_func": 0.0,
388
  "rewards/strict_format_reward_func": 0.0,
389
- "rewards/xmlcount_reward_func": -0.007607143372297287,
390
  "step": 40
391
  },
392
  {
393
- "completion_length": 370.59375,
394
  "epoch": 5.258064516129032,
395
- "grad_norm": 1.900159478187561,
396
- "kl": 7.740367414044158e-05,
397
  "learning_rate": 1.7017461746600504e-07,
398
  "loss": 0.0,
399
- "reward": 0.4386673276312649,
400
- "reward_std": 0.3562059886753559,
401
  "rewards/concensus_correctness_reward_func": 0.0,
402
  "rewards/consensus_reward_func": 0.0,
403
  "rewards/cumulative_reward_2": 0.0,
404
  "rewards/final_correctness_reward_func": 0.0,
405
- "rewards/question_recreation_reward_func": 0.2130110774305649,
406
  "rewards/soft_format_reward_func": 0.0,
407
  "rewards/strict_format_reward_func": 0.0,
408
- "rewards/xmlcount_reward_func": 0.22565624909475446,
409
  "step": 42
410
  },
411
  {
412
- "completion_length": 575.25,
413
  "epoch": 5.516129032258064,
414
- "grad_norm": 1.1061934232711792,
415
- "kl": 2.8744487849508005e-05,
416
  "learning_rate": 1.4502772460993384e-07,
417
  "loss": 0.0,
418
- "reward": 0.04084427189081907,
419
- "reward_std": 0.2996016829274595,
420
  "rewards/concensus_correctness_reward_func": 0.0,
421
  "rewards/consensus_reward_func": 0.0,
422
  "rewards/cumulative_reward_2": 0.0,
423
  "rewards/final_correctness_reward_func": 0.0,
424
- "rewards/question_recreation_reward_func": 0.08375052409246564,
425
  "rewards/soft_format_reward_func": 0.0,
426
  "rewards/strict_format_reward_func": 0.0,
427
- "rewards/xmlcount_reward_func": -0.04290625033900142,
428
  "step": 44
429
  },
430
  {
431
- "completion_length": 512.59375,
432
  "epoch": 5.774193548387097,
433
- "grad_norm": 31.07492446899414,
434
- "kl": 0.0005856987711467809,
435
  "learning_rate": 1.2111153570574453e-07,
436
  "loss": 0.0,
437
- "reward": 0.2757360441610217,
438
- "reward_std": 0.34823702834546566,
439
  "rewards/concensus_correctness_reward_func": 0.0,
440
  "rewards/consensus_reward_func": 0.0,
441
  "rewards/cumulative_reward_2": 0.0,
442
  "rewards/final_correctness_reward_func": 0.0,
443
- "rewards/question_recreation_reward_func": 0.1400797851383686,
444
  "rewards/soft_format_reward_func": 0.0,
445
  "rewards/strict_format_reward_func": 0.0,
446
- "rewards/xmlcount_reward_func": 0.13565625250339508,
447
  "step": 46
448
  },
449
  {
450
- "completion_length": 368.89285714285717,
451
  "epoch": 6.0,
452
- "grad_norm": 1.8032582998275757,
453
- "kl": 7.638505134569382e-05,
454
  "learning_rate": 9.870644620155877e-08,
455
  "loss": 0.0,
456
- "reward": 0.19913670314209803,
457
- "reward_std": 0.5407656750508717,
458
- "rewards/concensus_correctness_reward_func": 0.008928571428571428,
459
  "rewards/consensus_reward_func": 0.0,
460
  "rewards/cumulative_reward_2": 0.0,
461
  "rewards/final_correctness_reward_func": 0.0,
462
- "rewards/question_recreation_reward_func": 0.24160099548420735,
463
  "rewards/soft_format_reward_func": 0.0,
464
  "rewards/strict_format_reward_func": 0.0,
465
- "rewards/xmlcount_reward_func": -0.05139286071062088,
466
  "step": 48
467
  },
468
  {
469
- "completion_length": 216.34375,
470
  "epoch": 6.258064516129032,
471
- "grad_norm": 3.3991875648498535,
472
- "kl": 0.0012653063204197679,
473
  "learning_rate": 7.807513528664414e-08,
474
  "loss": 0.0,
475
- "reward": 0.1986216944642365,
476
- "reward_std": 0.18251595762558281,
477
  "rewards/concensus_correctness_reward_func": 0.0,
478
  "rewards/consensus_reward_func": 0.0,
479
  "rewards/cumulative_reward_2": 0.0,
480
  "rewards/final_correctness_reward_func": 0.0,
481
- "rewards/question_recreation_reward_func": 0.15384044079110026,
482
  "rewards/soft_format_reward_func": 0.0,
483
  "rewards/strict_format_reward_func": 0.0,
484
- "rewards/xmlcount_reward_func": 0.044781249947845936,
485
  "step": 50
486
  },
487
  {
488
- "completion_length": 449.8125,
489
  "epoch": 6.516129032258064,
490
- "grad_norm": 2.6445517539978027,
491
- "kl": 0.00018877460297517246,
492
  "learning_rate": 5.945948621809091e-08,
493
  "loss": 0.0,
494
- "reward": 0.14643485844135284,
495
- "reward_std": 0.399348977021873,
496
- "rewards/concensus_correctness_reward_func": 0.0,
497
  "rewards/consensus_reward_func": 0.0,
498
  "rewards/cumulative_reward_2": 0.0,
499
  "rewards/final_correctness_reward_func": 0.0,
500
- "rewards/question_recreation_reward_func": 0.158841117285192,
501
  "rewards/soft_format_reward_func": 0.0,
502
  "rewards/strict_format_reward_func": 0.0,
503
- "rewards/xmlcount_reward_func": -0.01240625255741179,
504
  "step": 52
505
  },
506
  {
507
- "completion_length": 347.28125,
508
  "epoch": 6.774193548387097,
509
- "grad_norm": 1.4666146039962769,
510
- "kl": 0.0003983834214977833,
511
  "learning_rate": 4.3077750460777384e-08,
512
  "loss": 0.0,
513
- "reward": 0.19152415730059147,
514
- "reward_std": 0.27376102562993765,
515
  "rewards/concensus_correctness_reward_func": 0.0,
516
  "rewards/consensus_reward_func": 0.0,
517
  "rewards/cumulative_reward_2": 0.0,
518
  "rewards/final_correctness_reward_func": 0.0,
519
- "rewards/question_recreation_reward_func": 0.12736790860071778,
520
  "rewards/soft_format_reward_func": 0.0,
521
  "rewards/strict_format_reward_func": 0.0,
522
- "rewards/xmlcount_reward_func": 0.06415624869987369,
523
  "step": 54
524
  },
525
  {
526
- "completion_length": 498.32142857142856,
527
  "epoch": 7.0,
528
- "grad_norm": 1.52250337600708,
529
- "kl": 0.00023215323692836267,
530
  "learning_rate": 2.912198888849429e-08,
531
  "loss": 0.0,
532
- "reward": 0.12091957139117378,
533
- "reward_std": 0.2569341185901846,
534
  "rewards/concensus_correctness_reward_func": 0.0,
535
  "rewards/consensus_reward_func": 0.0,
536
  "rewards/cumulative_reward_2": 0.0,
537
  "rewards/final_correctness_reward_func": 0.0,
538
- "rewards/question_recreation_reward_func": 0.1385624294302293,
539
  "rewards/soft_format_reward_func": 0.0,
540
  "rewards/strict_format_reward_func": 0.0,
541
- "rewards/xmlcount_reward_func": -0.017642855311610868,
542
  "step": 56
543
  },
544
  {
545
- "completion_length": 290.09375,
546
  "epoch": 7.258064516129032,
547
- "grad_norm": 1.6776108741760254,
548
- "kl": 0.0008473875428762767,
549
  "learning_rate": 1.7755820045802144e-08,
550
  "loss": 0.0,
551
- "reward": 0.3240678207948804,
552
- "reward_std": 0.37887709541246295,
553
  "rewards/concensus_correctness_reward_func": 0.0,
554
  "rewards/consensus_reward_func": 0.0,
555
  "rewards/cumulative_reward_2": 0.0,
556
  "rewards/final_correctness_reward_func": 0.0,
557
- "rewards/question_recreation_reward_func": 0.15538030490279198,
558
  "rewards/soft_format_reward_func": 0.0,
559
  "rewards/strict_format_reward_func": 0.0,
560
- "rewards/xmlcount_reward_func": 0.16868749679997563,
561
  "step": 58
562
  },
563
  {
564
- "completion_length": 422.78125,
565
  "epoch": 7.516129032258064,
566
- "grad_norm": 4.0938944816589355,
567
- "kl": 7.57059717670927e-05,
568
  "learning_rate": 9.112501870194272e-09,
569
  "loss": 0.0,
570
- "reward": 0.16282579209655523,
571
- "reward_std": 0.3961476086406037,
572
  "rewards/concensus_correctness_reward_func": 0.0,
573
  "rewards/consensus_reward_func": 0.0,
574
  "rewards/cumulative_reward_2": 0.0,
575
  "rewards/final_correctness_reward_func": 0.0,
576
- "rewards/question_recreation_reward_func": 0.12666953651933,
577
  "rewards/soft_format_reward_func": 0.0,
578
  "rewards/strict_format_reward_func": 0.0,
579
- "rewards/xmlcount_reward_func": 0.0361562481848523,
580
  "step": 60
581
  },
582
  {
583
  "epoch": 7.516129032258064,
584
  "step": 60,
585
  "total_flos": 0.0,
586
- "train_loss": 2.2126714176451402e-07,
587
- "train_runtime": 2214.7036,
588
- "train_samples_per_second": 0.433,
589
- "train_steps_per_second": 0.027
590
  }
591
  ],
592
  "logging_steps": 2,
 
10
  "is_world_process_zero": true,
11
  "log_history": [
12
  {
13
+ "completion_length": 430.8125,
14
  "epoch": 0.25806451612903225,
15
  "grad_norm": NaN,
16
+ "kl": 0.00041734772025847633,
17
  "learning_rate": 2.5e-07,
18
  "loss": 0.0,
19
+ "reward": 0.11391110043041408,
20
+ "reward_std": 0.41625179536640644,
21
  "rewards/concensus_correctness_reward_func": 0.0,
22
  "rewards/consensus_reward_func": 0.0,
23
  "rewards/cumulative_reward_2": 0.0,
24
  "rewards/final_correctness_reward_func": 0.0,
25
+ "rewards/question_recreation_reward_func": 0.20462985103949904,
26
  "rewards/soft_format_reward_func": 0.0,
27
  "rewards/strict_format_reward_func": 0.0,
28
+ "rewards/xmlcount_reward_func": -0.09071875200606883,
29
  "step": 2
30
  },
31
  {
32
+ "completion_length": 343.34375,
33
  "epoch": 0.5161290322580645,
34
+ "grad_norm": 1.9656559228897095,
35
+ "kl": 0.00040763827485079673,
36
  "learning_rate": 5e-07,
37
  "loss": 0.0,
38
+ "reward": 0.08423173008486629,
39
+ "reward_std": 0.1950428435811773,
40
  "rewards/concensus_correctness_reward_func": 0.0,
41
  "rewards/consensus_reward_func": 0.0,
42
  "rewards/cumulative_reward_2": 0.0,
43
  "rewards/final_correctness_reward_func": 0.0,
44
+ "rewards/question_recreation_reward_func": 0.10804423363879323,
45
  "rewards/soft_format_reward_func": 0.0,
46
  "rewards/strict_format_reward_func": 0.0,
47
+ "rewards/xmlcount_reward_func": -0.02381250262260437,
48
  "step": 4
49
  },
50
  {
51
+ "completion_length": 426.71875,
52
  "epoch": 0.7741935483870968,
53
+ "grad_norm": 3.116727590560913,
54
+ "kl": 0.00013401095452536538,
55
+ "learning_rate": 4.985344892885899e-07,
56
  "loss": 0.0,
57
+ "reward": 0.2961328737437725,
58
+ "reward_std": 0.3188573829829693,
59
  "rewards/concensus_correctness_reward_func": 0.0,
60
  "rewards/consensus_reward_func": 0.0,
61
  "rewards/cumulative_reward_2": 0.0,
62
  "rewards/final_correctness_reward_func": 0.0,
63
+ "rewards/question_recreation_reward_func": 0.20203911093994975,
64
  "rewards/soft_format_reward_func": 0.0,
65
  "rewards/strict_format_reward_func": 0.0,
66
+ "rewards/xmlcount_reward_func": 0.09409375116229057,
67
  "step": 6
68
  },
69
  {
70
+ "completion_length": 363.85714285714283,
71
  "epoch": 1.0,
72
+ "grad_norm": 4.322468280792236,
73
+ "kl": 0.000584651980066805,
74
  "learning_rate": 4.967066306353815e-07,
75
  "loss": 0.0,
76
+ "reward": 0.28172825063977924,
77
+ "reward_std": 0.2894980385899544,
78
  "rewards/concensus_correctness_reward_func": 0.0,
79
  "rewards/consensus_reward_func": 0.0,
80
  "rewards/cumulative_reward_2": 0.0,
81
  "rewards/final_correctness_reward_func": 0.0,
82
+ "rewards/question_recreation_reward_func": 0.1669425368309021,
83
  "rewards/soft_format_reward_func": 0.0,
84
  "rewards/strict_format_reward_func": 0.0,
85
+ "rewards/xmlcount_reward_func": 0.11478571407496929,
86
  "step": 8
87
  },
88
  {
89
+ "completion_length": 375.15625,
90
  "epoch": 1.2580645161290323,
91
+ "grad_norm": 4.052526473999023,
92
+ "kl": 0.00137491720761318,
93
+ "learning_rate": 4.941551389275217e-07,
94
  "loss": 0.0,
95
+ "reward": 0.1756317124236375,
96
+ "reward_std": 0.13710087712388486,
97
  "rewards/concensus_correctness_reward_func": 0.0,
98
  "rewards/consensus_reward_func": 0.0,
99
  "rewards/cumulative_reward_2": 0.0,
100
  "rewards/final_correctness_reward_func": 0.0,
101
+ "rewards/question_recreation_reward_func": 0.10741296270862222,
102
  "rewards/soft_format_reward_func": 0.0,
103
  "rewards/strict_format_reward_func": 0.0,
104
+ "rewards/xmlcount_reward_func": 0.06821874901652336,
105
  "step": 10
106
  },
107
  {
108
+ "completion_length": 527.28125,
109
  "epoch": 1.5161290322580645,
110
+ "grad_norm": 1.0876020193099976,
111
+ "kl": 7.662167183752899e-05,
112
  "learning_rate": 4.869132927957006e-07,
113
  "loss": 0.0,
114
+ "reward": 0.05637967900838703,
115
+ "reward_std": 0.29673084686510265,
116
  "rewards/concensus_correctness_reward_func": 0.0,
117
  "rewards/consensus_reward_func": 0.0,
118
  "rewards/cumulative_reward_2": 0.0,
119
  "rewards/final_correctness_reward_func": 0.0,
120
+ "rewards/question_recreation_reward_func": 0.08856716414447874,
121
  "rewards/soft_format_reward_func": 0.0,
122
  "rewards/strict_format_reward_func": 0.0,
123
+ "rewards/xmlcount_reward_func": -0.03218749677762389,
124
  "step": 12
125
  },
126
  {
127
+ "completion_length": 420.4375,
128
  "epoch": 1.7741935483870968,
129
+ "grad_norm": 1.930647373199463,
130
+ "kl": 0.00019077809974987758,
131
  "learning_rate": 4.768938549177392e-07,
132
  "loss": 0.0,
133
+ "reward": 0.32431007781997323,
134
+ "reward_std": 0.3349539686460048,
135
  "rewards/concensus_correctness_reward_func": 0.0,
136
  "rewards/consensus_reward_func": 0.0,
137
  "rewards/cumulative_reward_2": 0.0,
138
  "rewards/final_correctness_reward_func": 0.0,
139
+ "rewards/question_recreation_reward_func": 0.19974758382886648,
140
  "rewards/soft_format_reward_func": 0.0,
141
  "rewards/strict_format_reward_func": 0.0,
142
+ "rewards/xmlcount_reward_func": 0.12456249631941319,
143
  "step": 14
144
  },
145
  {
146
+ "completion_length": 235.67857142857142,
147
  "epoch": 2.0,
148
+ "grad_norm": 25.613550186157227,
149
+ "kl": 0.001633655218029162,
150
+ "learning_rate": 4.708780111115057e-07,
151
+ "loss": 0.0,
152
+ "reward": 0.187016135347741,
153
+ "reward_std": 0.25221954605409075,
154
  "rewards/concensus_correctness_reward_func": 0.0,
155
  "rewards/consensus_reward_func": 0.0,
156
  "rewards/cumulative_reward_2": 0.0,
157
  "rewards/final_correctness_reward_func": 0.0,
158
+ "rewards/question_recreation_reward_func": 0.11437327787280083,
159
  "rewards/soft_format_reward_func": 0.0,
160
  "rewards/strict_format_reward_func": 0.0,
161
+ "rewards/xmlcount_reward_func": 0.07264285454792636,
162
  "step": 16
163
  },
164
  {
165
+ "completion_length": 318.28125,
166
  "epoch": 2.258064516129032,
167
+ "grad_norm": 3.3770768642425537,
168
+ "kl": 0.0005925071802721504,
169
  "learning_rate": 4.5692224953922265e-07,
170
  "loss": 0.0,
171
+ "reward": 0.2636642586439848,
172
+ "reward_std": 0.3803391242399812,
173
  "rewards/concensus_correctness_reward_func": 0.0,
174
  "rewards/consensus_reward_func": 0.0,
175
  "rewards/cumulative_reward_2": 0.0,
176
  "rewards/final_correctness_reward_func": 0.0,
177
+ "rewards/question_recreation_reward_func": 0.13447676633950323,
178
  "rewards/soft_format_reward_func": 0.0,
179
  "rewards/strict_format_reward_func": 0.0,
180
+ "rewards/xmlcount_reward_func": 0.12918749847449362,
181
  "step": 18
182
  },
183
  {
184
+ "completion_length": 371.59375,
185
  "epoch": 2.5161290322580645,
186
+ "grad_norm": 1.974790096282959,
187
+ "kl": 0.0002036209966718161,
188
  "learning_rate": 4.405405137819091e-07,
189
  "loss": 0.0,
190
+ "reward": 0.10157998092472553,
191
+ "reward_std": 0.10903631476685405,
192
  "rewards/concensus_correctness_reward_func": 0.0,
193
  "rewards/consensus_reward_func": 0.0,
194
  "rewards/cumulative_reward_2": 0.0,
195
  "rewards/final_correctness_reward_func": 0.0,
196
+ "rewards/question_recreation_reward_func": 0.06636122916825116,
197
  "rewards/soft_format_reward_func": 0.0,
198
  "rewards/strict_format_reward_func": 0.0,
199
+ "rewards/xmlcount_reward_func": 0.035218749195337296,
200
  "step": 20
201
  },
202
  {
203
+ "completion_length": 279.71875,
204
  "epoch": 2.774193548387097,
205
+ "grad_norm": 26.12720489501953,
206
+ "kl": 0.0019963133127021138,
207
  "learning_rate": 4.219248647133558e-07,
208
  "loss": 0.0,
209
+ "reward": 0.35939981369301677,
210
+ "reward_std": 0.5500808532815427,
211
  "rewards/concensus_correctness_reward_func": 0.0,
212
  "rewards/consensus_reward_func": 0.0,
213
  "rewards/cumulative_reward_2": 0.0,
214
  "rewards/final_correctness_reward_func": 0.0,
215
+ "rewards/question_recreation_reward_func": 0.25952482083812356,
216
  "rewards/soft_format_reward_func": 0.0,
217
  "rewards/strict_format_reward_func": 0.0,
218
+ "rewards/xmlcount_reward_func": 0.09987499937415123,
219
  "step": 22
220
  },
221
  {
222
+ "completion_length": 383.64285714285717,
223
  "epoch": 3.0,
224
+ "grad_norm": 0.9543819427490234,
225
+ "kl": 0.0006048666277170664,
226
  "learning_rate": 4.012935537984413e-07,
227
  "loss": 0.0,
228
+ "reward": 0.22724560940904276,
229
+ "reward_std": 0.20333547570875712,
230
  "rewards/concensus_correctness_reward_func": 0.0,
231
  "rewards/consensus_reward_func": 0.0,
232
  "rewards/cumulative_reward_2": 0.0,
233
  "rewards/final_correctness_reward_func": 0.0,
234
+ "rewards/question_recreation_reward_func": 0.18638846703938075,
235
  "rewards/soft_format_reward_func": 0.0,
236
  "rewards/strict_format_reward_func": 0.0,
237
+ "rewards/xmlcount_reward_func": 0.04085714370012283,
238
  "step": 24
239
  },
240
  {
241
+ "completion_length": 381.96875,
242
  "epoch": 3.258064516129032,
243
+ "grad_norm": 3.211329698562622,
244
+ "kl": 0.00012533261292446696,
245
  "learning_rate": 3.788884642942554e-07,
246
  "loss": 0.0,
247
+ "reward": 0.19361284282058477,
248
+ "reward_std": 0.24643944506533444,
249
  "rewards/concensus_correctness_reward_func": 0.0,
250
  "rewards/consensus_reward_func": 0.0,
251
  "rewards/cumulative_reward_2": 0.0,
252
  "rewards/final_correctness_reward_func": 0.0,
253
+ "rewards/question_recreation_reward_func": 0.17839408991858363,
254
  "rewards/soft_format_reward_func": 0.0,
255
  "rewards/strict_format_reward_func": 0.0,
256
+ "rewards/xmlcount_reward_func": 0.015218751970678568,
257
  "step": 26
258
  },
259
  {
260
+ "completion_length": 412.5,
261
  "epoch": 3.5161290322580645,
262
+ "grad_norm": 3.394382953643799,
263
+ "kl": 0.00010090849150401482,
264
  "learning_rate": 3.5497227539006614e-07,
265
  "loss": 0.0,
266
+ "reward": 0.2241632086224854,
267
+ "reward_std": 0.2526557689998299,
268
+ "rewards/concensus_correctness_reward_func": 0.013875000178813934,
269
  "rewards/consensus_reward_func": 0.0,
270
  "rewards/cumulative_reward_2": 0.0,
271
  "rewards/final_correctness_reward_func": 0.0,
272
+ "rewards/question_recreation_reward_func": 0.17597571481019258,
273
  "rewards/soft_format_reward_func": 0.0,
274
  "rewards/strict_format_reward_func": 0.0,
275
+ "rewards/xmlcount_reward_func": 0.03431250061839819,
276
  "step": 28
277
  },
278
  {
279
+ "completion_length": 284.59375,
280
  "epoch": 3.774193548387097,
281
+ "grad_norm": 2.9376416206359863,
282
+ "kl": 0.0008606196147411538,
283
  "learning_rate": 3.2982538253399494e-07,
284
  "loss": 0.0,
285
+ "reward": 0.0774760153144598,
286
+ "reward_std": 0.1906891551334411,
287
  "rewards/concensus_correctness_reward_func": 0.0,
288
  "rewards/consensus_reward_func": 0.0,
289
  "rewards/cumulative_reward_2": 0.0,
290
  "rewards/final_correctness_reward_func": 0.0,
291
+ "rewards/question_recreation_reward_func": 0.09122601104900241,
292
  "rewards/soft_format_reward_func": 0.0,
293
  "rewards/strict_format_reward_func": 0.0,
294
+ "rewards/xmlcount_reward_func": -0.01374999899417162,
295
  "step": 30
296
  },
297
  {
298
+ "completion_length": 549.1428571428571,
299
  "epoch": 4.0,
300
+ "grad_norm": 0.9819178581237793,
301
+ "kl": 3.5185036722396035e-05,
302
  "learning_rate": 3.0374261005275605e-07,
303
  "loss": 0.0,
304
+ "reward": 0.25383392149316414,
305
+ "reward_std": 0.30834708123334814,
306
  "rewards/concensus_correctness_reward_func": 0.0,
307
  "rewards/consensus_reward_func": 0.0,
308
  "rewards/cumulative_reward_2": 0.0,
309
  "rewards/final_correctness_reward_func": 0.0,
310
+ "rewards/question_recreation_reward_func": 0.11165534212653126,
311
  "rewards/soft_format_reward_func": 0.0,
312
  "rewards/strict_format_reward_func": 0.0,
313
+ "rewards/xmlcount_reward_func": 0.14217857164996012,
314
  "step": 32
315
  },
316
  {
317
+ "completion_length": 366.59375,
318
  "epoch": 4.258064516129032,
319
+ "grad_norm": 2.5270180702209473,
320
+ "kl": 0.004282280417555739,
321
  "learning_rate": 2.770297546059854e-07,
322
  "loss": 0.0,
323
+ "reward": 0.06346084550023079,
324
+ "reward_std": 0.28176579938735813,
325
  "rewards/concensus_correctness_reward_func": 0.0,
326
  "rewards/consensus_reward_func": 0.0,
327
  "rewards/cumulative_reward_2": 0.0,
328
  "rewards/final_correctness_reward_func": 0.0,
329
+ "rewards/question_recreation_reward_func": 0.12233584048226476,
330
  "rewards/soft_format_reward_func": 0.0,
331
  "rewards/strict_format_reward_func": 0.0,
332
+ "rewards/xmlcount_reward_func": -0.0588750010356307,
333
  "step": 34
334
  },
335
  {
336
+ "completion_length": 411.375,
337
  "epoch": 4.516129032258064,
338
+ "grad_norm": 4.21773099899292,
339
+ "kl": 8.090390974757611e-05,
340
  "learning_rate": 2.5e-07,
341
  "loss": 0.0,
342
+ "reward": 0.11983968829736114,
343
+ "reward_std": 0.17383667826652527,
344
  "rewards/concensus_correctness_reward_func": 0.0,
345
  "rewards/consensus_reward_func": 0.0,
346
  "rewards/cumulative_reward_2": 0.0,
347
  "rewards/final_correctness_reward_func": 0.0,
348
+ "rewards/question_recreation_reward_func": 0.11633968772366643,
349
  "rewards/soft_format_reward_func": 0.0,
350
  "rewards/strict_format_reward_func": 0.0,
351
+ "rewards/xmlcount_reward_func": 0.0035000010393559933,
352
  "step": 36
353
  },
354
  {
355
+ "completion_length": 329.28125,
356
  "epoch": 4.774193548387097,
357
+ "grad_norm": 18.660564422607422,
358
+ "kl": 0.0010729862383414002,
359
  "learning_rate": 2.229702453940146e-07,
360
  "loss": 0.0,
361
+ "reward": 0.22536780731752515,
362
+ "reward_std": 0.22597291774582118,
363
  "rewards/concensus_correctness_reward_func": 0.0,
364
  "rewards/consensus_reward_func": 0.0,
365
  "rewards/cumulative_reward_2": 0.0,
366
  "rewards/final_correctness_reward_func": 0.0,
367
+ "rewards/question_recreation_reward_func": 0.13943030545488,
368
  "rewards/soft_format_reward_func": 0.0,
369
  "rewards/strict_format_reward_func": 0.0,
370
+ "rewards/xmlcount_reward_func": 0.0859375,
371
  "step": 38
372
  },
373
  {
374
+ "completion_length": 466.82142857142856,
375
  "epoch": 5.0,
376
+ "grad_norm": 30.90364646911621,
377
+ "kl": 0.0050324718344459795,
378
  "learning_rate": 1.9625738994724398e-07,
379
  "loss": 0.0,
380
+ "reward": -0.09624448684709412,
381
+ "reward_std": 0.43322019491876873,
382
  "rewards/concensus_correctness_reward_func": 0.0,
383
  "rewards/consensus_reward_func": 0.0,
384
  "rewards/cumulative_reward_2": 0.0,
385
  "rewards/final_correctness_reward_func": 0.0,
386
+ "rewards/question_recreation_reward_func": 0.08754122070968151,
387
  "rewards/soft_format_reward_func": 0.0,
388
  "rewards/strict_format_reward_func": 0.0,
389
+ "rewards/xmlcount_reward_func": -0.18378571208034242,
390
  "step": 40
391
  },
392
  {
393
+ "completion_length": 434.0625,
394
  "epoch": 5.258064516129032,
395
+ "grad_norm": 39.12836837768555,
396
+ "kl": 0.005596646230287661,
397
  "learning_rate": 1.7017461746600504e-07,
398
  "loss": 0.0,
399
+ "reward": 0.13575120829045773,
400
+ "reward_std": 0.13530697469832376,
401
  "rewards/concensus_correctness_reward_func": 0.0,
402
  "rewards/consensus_reward_func": 0.0,
403
  "rewards/cumulative_reward_2": 0.0,
404
  "rewards/final_correctness_reward_func": 0.0,
405
+ "rewards/question_recreation_reward_func": 0.1172824571840465,
406
  "rewards/soft_format_reward_func": 0.0,
407
  "rewards/strict_format_reward_func": 0.0,
408
+ "rewards/xmlcount_reward_func": 0.01846875064074993,
409
  "step": 42
410
  },
411
  {
412
+ "completion_length": 394.4375,
413
  "epoch": 5.516129032258064,
414
+ "grad_norm": 20.784292221069336,
415
+ "kl": 0.0022700289155181963,
416
  "learning_rate": 1.4502772460993384e-07,
417
  "loss": 0.0,
418
+ "reward": 0.11197976651601493,
419
+ "reward_std": 0.5696409860393032,
420
  "rewards/concensus_correctness_reward_func": 0.0,
421
  "rewards/consensus_reward_func": 0.0,
422
  "rewards/cumulative_reward_2": 0.0,
423
  "rewards/final_correctness_reward_func": 0.0,
424
+ "rewards/question_recreation_reward_func": 0.13532351609319448,
425
  "rewards/soft_format_reward_func": 0.0,
426
  "rewards/strict_format_reward_func": 0.0,
427
+ "rewards/xmlcount_reward_func": -0.02334374561905861,
428
  "step": 44
429
  },
430
  {
431
+ "completion_length": 360.03125,
432
  "epoch": 5.774193548387097,
433
+ "grad_norm": 4.866372585296631,
434
+ "kl": 0.000709792191400993,
435
  "learning_rate": 1.2111153570574453e-07,
436
  "loss": 0.0,
437
+ "reward": 0.1314556710422039,
438
+ "reward_std": 0.27511434676125646,
439
  "rewards/concensus_correctness_reward_func": 0.0,
440
  "rewards/consensus_reward_func": 0.0,
441
  "rewards/cumulative_reward_2": 0.0,
442
  "rewards/final_correctness_reward_func": 0.0,
443
+ "rewards/question_recreation_reward_func": 0.1430806666612625,
444
  "rewards/soft_format_reward_func": 0.0,
445
  "rewards/strict_format_reward_func": 0.0,
446
+ "rewards/xmlcount_reward_func": -0.011624999344348907,
447
  "step": 46
448
  },
449
  {
450
+ "completion_length": 201.39285714285714,
451
  "epoch": 6.0,
452
+ "grad_norm": 2.460200786590576,
453
+ "kl": 0.0008755299386393744,
454
  "learning_rate": 9.870644620155877e-08,
455
  "loss": 0.0,
456
+ "reward": 0.26238812506198883,
457
+ "reward_std": 0.20772267784923315,
458
+ "rewards/concensus_correctness_reward_func": 0.0,
459
  "rewards/consensus_reward_func": 0.0,
460
  "rewards/cumulative_reward_2": 0.0,
461
  "rewards/final_correctness_reward_func": 0.0,
462
+ "rewards/question_recreation_reward_func": 0.23092382933412278,
463
  "rewards/soft_format_reward_func": 0.0,
464
  "rewards/strict_format_reward_func": 0.0,
465
+ "rewards/xmlcount_reward_func": 0.031464285616363795,
466
  "step": 48
467
  },
468
  {
469
+ "completion_length": 346.40625,
470
  "epoch": 6.258064516129032,
471
+ "grad_norm": 21.511545181274414,
472
+ "kl": 0.0017042956070554283,
473
  "learning_rate": 7.807513528664414e-08,
474
  "loss": 0.0,
475
+ "reward": 0.14076933730393648,
476
+ "reward_std": 0.16657349606975913,
477
  "rewards/concensus_correctness_reward_func": 0.0,
478
  "rewards/consensus_reward_func": 0.0,
479
  "rewards/cumulative_reward_2": 0.0,
480
  "rewards/final_correctness_reward_func": 0.0,
481
+ "rewards/question_recreation_reward_func": 0.07820683810859919,
482
  "rewards/soft_format_reward_func": 0.0,
483
  "rewards/strict_format_reward_func": 0.0,
484
+ "rewards/xmlcount_reward_func": 0.06256250012665987,
485
  "step": 50
486
  },
487
  {
488
+ "completion_length": 325.4375,
489
  "epoch": 6.516129032258064,
490
+ "grad_norm": 1.8207321166992188,
491
+ "kl": 0.0018291326173311973,
492
  "learning_rate": 5.945948621809091e-08,
493
  "loss": 0.0,
494
+ "reward": 0.19795077852904797,
495
+ "reward_std": 0.22306197369471192,
496
+ "rewards/concensus_correctness_reward_func": 0.013875000178813934,
497
  "rewards/consensus_reward_func": 0.0,
498
  "rewards/cumulative_reward_2": 0.0,
499
  "rewards/final_correctness_reward_func": 0.0,
500
+ "rewards/question_recreation_reward_func": 0.16435701958835125,
501
  "rewards/soft_format_reward_func": 0.0,
502
  "rewards/strict_format_reward_func": 0.0,
503
+ "rewards/xmlcount_reward_func": 0.01971874851733446,
504
  "step": 52
505
  },
506
  {
507
+ "completion_length": 414.5625,
508
  "epoch": 6.774193548387097,
509
+ "grad_norm": 3.8619086742401123,
510
+ "kl": 0.0004780482821615806,
511
  "learning_rate": 4.3077750460777384e-08,
512
  "loss": 0.0,
513
+ "reward": 0.09389630286023021,
514
+ "reward_std": 0.19037447159644216,
515
  "rewards/concensus_correctness_reward_func": 0.0,
516
  "rewards/consensus_reward_func": 0.0,
517
  "rewards/cumulative_reward_2": 0.0,
518
  "rewards/final_correctness_reward_func": 0.0,
519
+ "rewards/question_recreation_reward_func": 0.11933380062691867,
520
  "rewards/soft_format_reward_func": 0.0,
521
  "rewards/strict_format_reward_func": 0.0,
522
+ "rewards/xmlcount_reward_func": -0.02543749986216426,
523
  "step": 54
524
  },
525
  {
526
+ "completion_length": 342.5,
527
  "epoch": 7.0,
528
+ "grad_norm": 1.9887802600860596,
529
+ "kl": 0.0015315435006674047,
530
  "learning_rate": 2.912198888849429e-08,
531
  "loss": 0.0,
532
+ "reward": 0.1833051379237856,
533
+ "reward_std": 0.16237848784242356,
534
  "rewards/concensus_correctness_reward_func": 0.0,
535
  "rewards/consensus_reward_func": 0.0,
536
  "rewards/cumulative_reward_2": 0.0,
537
  "rewards/final_correctness_reward_func": 0.0,
538
+ "rewards/question_recreation_reward_func": 0.11926942531551633,
539
  "rewards/soft_format_reward_func": 0.0,
540
  "rewards/strict_format_reward_func": 0.0,
541
+ "rewards/xmlcount_reward_func": 0.06403571314045362,
542
  "step": 56
543
  },
544
  {
545
+ "completion_length": 335.1875,
546
  "epoch": 7.258064516129032,
547
+ "grad_norm": 2.110353469848633,
548
+ "kl": 0.00010102857913807384,
549
  "learning_rate": 1.7755820045802144e-08,
550
  "loss": 0.0,
551
+ "reward": 0.14824581565335393,
552
+ "reward_std": 0.22165359160862863,
553
  "rewards/concensus_correctness_reward_func": 0.0,
554
  "rewards/consensus_reward_func": 0.0,
555
  "rewards/cumulative_reward_2": 0.0,
556
  "rewards/final_correctness_reward_func": 0.0,
557
+ "rewards/question_recreation_reward_func": 0.16133957030251622,
558
  "rewards/soft_format_reward_func": 0.0,
559
  "rewards/strict_format_reward_func": 0.0,
560
+ "rewards/xmlcount_reward_func": -0.013093750923871994,
561
  "step": 58
562
  },
563
  {
564
+ "completion_length": 431.1875,
565
  "epoch": 7.516129032258064,
566
+ "grad_norm": 2.8052024841308594,
567
+ "kl": 0.001484481835632323,
568
  "learning_rate": 9.112501870194272e-09,
569
  "loss": 0.0,
570
+ "reward": 0.3731462989235297,
571
+ "reward_std": 0.42140341037884355,
572
  "rewards/concensus_correctness_reward_func": 0.0,
573
  "rewards/consensus_reward_func": 0.0,
574
  "rewards/cumulative_reward_2": 0.0,
575
  "rewards/final_correctness_reward_func": 0.0,
576
+ "rewards/question_recreation_reward_func": 0.20230254053603858,
577
  "rewards/soft_format_reward_func": 0.0,
578
  "rewards/strict_format_reward_func": 0.0,
579
+ "rewards/xmlcount_reward_func": 0.17084375163540244,
580
  "step": 60
581
  },
582
  {
583
  "epoch": 7.516129032258064,
584
  "step": 60,
585
  "total_flos": 0.0,
586
+ "train_loss": 1.1757511778398565e-06,
587
+ "train_runtime": 2092.798,
588
+ "train_samples_per_second": 0.459,
589
+ "train_steps_per_second": 0.029
590
  }
591
  ],
592
  "logging_steps": 2,