File size: 10,910 Bytes
51b1f39
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9843342036553526,
  "eval_steps": 100,
  "global_step": 190,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 2.6315789473684213e-07,
      "logits/chosen": 0.8826487064361572,
      "logits/rejected": 0.921362042427063,
      "logps/chosen": -36.58121871948242,
      "logps/rejected": -54.902320861816406,
      "loss": 0.01,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.1,
      "learning_rate": 2.631578947368421e-06,
      "logits/chosen": 0.8915393352508545,
      "logits/rejected": 0.8742258548736572,
      "logps/chosen": -87.77196502685547,
      "logps/rejected": -96.38507843017578,
      "loss": 0.0101,
      "rewards/accuracies": 0.2569444477558136,
      "rewards/chosen": 0.0003006549668498337,
      "rewards/margins": 0.0004423653008416295,
      "rewards/rejected": -0.00014171031943988055,
      "step": 10
    },
    {
      "epoch": 0.21,
      "learning_rate": 4.999578104083307e-06,
      "logits/chosen": 0.7802013158798218,
      "logits/rejected": 0.8469624519348145,
      "logps/chosen": -91.75413513183594,
      "logps/rejected": -85.1173095703125,
      "loss": 0.01,
      "rewards/accuracies": 0.24375000596046448,
      "rewards/chosen": 3.31846640619915e-05,
      "rewards/margins": -0.00015664812235627323,
      "rewards/rejected": 0.00018983279005624354,
      "step": 20
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.949122667718935e-06,
      "logits/chosen": 0.8652933835983276,
      "logits/rejected": 0.848902702331543,
      "logps/chosen": -85.29698944091797,
      "logps/rejected": -78.0544204711914,
      "loss": 0.0099,
      "rewards/accuracies": 0.28125,
      "rewards/chosen": 0.0002226830692961812,
      "rewards/margins": 0.00038711068918928504,
      "rewards/rejected": -0.00016442763444501907,
      "step": 30
    },
    {
      "epoch": 0.42,
      "learning_rate": 4.8162351680370046e-06,
      "logits/chosen": 0.7649837136268616,
      "logits/rejected": 0.841802716255188,
      "logps/chosen": -122.76881408691406,
      "logps/rejected": -108.8086166381836,
      "loss": 0.01,
      "rewards/accuracies": 0.26875001192092896,
      "rewards/chosen": -0.000362707010935992,
      "rewards/margins": -7.97106167738093e-06,
      "rewards/rejected": -0.0003547359665390104,
      "step": 40
    },
    {
      "epoch": 0.52,
      "learning_rate": 4.605388304968915e-06,
      "logits/chosen": 0.8386613130569458,
      "logits/rejected": 0.8677732348442078,
      "logps/chosen": -65.77490997314453,
      "logps/rejected": -71.66779327392578,
      "loss": 0.01,
      "rewards/accuracies": 0.21250000596046448,
      "rewards/chosen": -0.00019812444224953651,
      "rewards/margins": -0.00015129637904465199,
      "rewards/rejected": -4.682801591116004e-05,
      "step": 50
    },
    {
      "epoch": 0.63,
      "learning_rate": 4.323678718546552e-06,
      "logits/chosen": 0.8152744174003601,
      "logits/rejected": 0.8866288065910339,
      "logps/chosen": -117.6309814453125,
      "logps/rejected": -110.9274673461914,
      "loss": 0.0101,
      "rewards/accuracies": 0.26875001192092896,
      "rewards/chosen": 2.0224386389600113e-05,
      "rewards/margins": -1.7978531104745343e-05,
      "rewards/rejected": 3.820297933998518e-05,
      "step": 60
    },
    {
      "epoch": 0.73,
      "learning_rate": 3.980588131662451e-06,
      "logits/chosen": 0.8049997091293335,
      "logits/rejected": 0.8617580533027649,
      "logps/chosen": -82.61628723144531,
      "logps/rejected": -83.92156982421875,
      "loss": 0.01,
      "rewards/accuracies": 0.3187499940395355,
      "rewards/chosen": 0.00046920054592192173,
      "rewards/margins": 0.000604915083386004,
      "rewards/rejected": -0.00013571445015259087,
      "step": 70
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.5876642162051833e-06,
      "logits/chosen": 0.7660447955131531,
      "logits/rejected": 0.8296969532966614,
      "logps/chosen": -105.82564544677734,
      "logps/rejected": -104.93913269042969,
      "loss": 0.01,
      "rewards/accuracies": 0.23749999701976776,
      "rewards/chosen": -0.0001158899613074027,
      "rewards/margins": -0.00039075990207493305,
      "rewards/rejected": 0.0002748699625954032,
      "step": 80
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.1581319239114983e-06,
      "logits/chosen": 0.8339036107063293,
      "logits/rejected": 0.8548374176025391,
      "logps/chosen": -77.63902282714844,
      "logps/rejected": -90.03132629394531,
      "loss": 0.0099,
      "rewards/accuracies": 0.36250001192092896,
      "rewards/chosen": 0.0005246674409136176,
      "rewards/margins": 0.0012112573022022843,
      "rewards/rejected": -0.0006865898030810058,
      "step": 90
    },
    {
      "epoch": 1.04,
      "learning_rate": 2.7064483636808314e-06,
      "logits/chosen": 0.7685378789901733,
      "logits/rejected": 0.8903138041496277,
      "logps/chosen": -91.08631896972656,
      "logps/rejected": -101.58308410644531,
      "loss": 0.01,
      "rewards/accuracies": 0.29374998807907104,
      "rewards/chosen": -0.00034642056562006474,
      "rewards/margins": 0.00024903417215682566,
      "rewards/rejected": -0.0005954547086730599,
      "step": 100
    },
    {
      "epoch": 1.04,
      "eval_logits/chosen": 0.8011398911476135,
      "eval_logits/rejected": 0.8187842965126038,
      "eval_logps/chosen": -91.76709747314453,
      "eval_logps/rejected": -94.26233673095703,
      "eval_loss": 0.009981811977922916,
      "eval_rewards/accuracies": 0.25,
      "eval_rewards/chosen": -0.00039719167398288846,
      "eval_rewards/margins": 0.00025612558238208294,
      "eval_rewards/rejected": -0.0006533172563649714,
      "eval_runtime": 274.2663,
      "eval_samples_per_second": 7.292,
      "eval_steps_per_second": 0.456,
      "step": 100
    },
    {
      "epoch": 1.15,
      "learning_rate": 2.2478162071993296e-06,
      "logits/chosen": 0.8606807589530945,
      "logits/rejected": 0.9067083597183228,
      "logps/chosen": -103.346435546875,
      "logps/rejected": -102.74019622802734,
      "loss": 0.0099,
      "rewards/accuracies": 0.3125,
      "rewards/chosen": 0.0002891735057346523,
      "rewards/margins": 0.0007119966903701425,
      "rewards/rejected": -0.00042282306822016835,
      "step": 110
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.797672000566077e-06,
      "logits/chosen": 0.8887661099433899,
      "logits/rejected": 0.8524330258369446,
      "logps/chosen": -92.60103607177734,
      "logps/rejected": -70.87162780761719,
      "loss": 0.0099,
      "rewards/accuracies": 0.2874999940395355,
      "rewards/chosen": 0.00012959113519173115,
      "rewards/margins": 0.0010213626082986593,
      "rewards/rejected": -0.0008917713421396911,
      "step": 120
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.3711666042227772e-06,
      "logits/chosen": 0.8812958598136902,
      "logits/rejected": 0.9191433191299438,
      "logps/chosen": -99.47935485839844,
      "logps/rejected": -102.90433502197266,
      "loss": 0.0099,
      "rewards/accuracies": 0.33125001192092896,
      "rewards/chosen": -0.00027921958826482296,
      "rewards/margins": 0.00039964643656276166,
      "rewards/rejected": -0.0006788660539314151,
      "step": 130
    },
    {
      "epoch": 1.46,
      "learning_rate": 9.826552484321086e-07,
      "logits/chosen": 0.8100296258926392,
      "logits/rejected": 0.8721168637275696,
      "logps/chosen": -98.98599243164062,
      "logps/rejected": -95.76078033447266,
      "loss": 0.0099,
      "rewards/accuracies": 0.29374998807907104,
      "rewards/chosen": -0.0003388571203686297,
      "rewards/margins": 0.0009376562084071338,
      "rewards/rejected": -0.0012765133287757635,
      "step": 140
    },
    {
      "epoch": 1.57,
      "learning_rate": 6.452143679117965e-07,
      "logits/chosen": 0.8696478009223938,
      "logits/rejected": 0.884810745716095,
      "logps/chosen": -74.74392700195312,
      "logps/rejected": -80.85291290283203,
      "loss": 0.0099,
      "rewards/accuracies": 0.2874999940395355,
      "rewards/chosen": -3.058182119275443e-05,
      "rewards/margins": 0.0010159575613215566,
      "rewards/rejected": -0.0010465392842888832,
      "step": 150
    },
    {
      "epoch": 1.67,
      "learning_rate": 3.7020147790418266e-07,
      "logits/chosen": 0.8866029977798462,
      "logits/rejected": 0.8834725618362427,
      "logps/chosen": -105.7184066772461,
      "logps/rejected": -99.87519836425781,
      "loss": 0.0099,
      "rewards/accuracies": 0.28125,
      "rewards/chosen": -0.0007261586142703891,
      "rewards/margins": 2.431169559713453e-05,
      "rewards/rejected": -0.000750470208004117,
      "step": 160
    },
    {
      "epoch": 1.78,
      "learning_rate": 1.6687290528135725e-07,
      "logits/chosen": 0.7714171409606934,
      "logits/rejected": 0.8409671783447266,
      "logps/chosen": -102.75898742675781,
      "logps/rejected": -96.38053894042969,
      "loss": 0.0099,
      "rewards/accuracies": 0.3375000059604645,
      "rewards/chosen": -0.0006845382740721107,
      "rewards/margins": 0.00033002972486428916,
      "rewards/rejected": -0.0010145680280402303,
      "step": 170
    },
    {
      "epoch": 1.88,
      "learning_rate": 4.207224101311247e-08,
      "logits/chosen": 0.8822728395462036,
      "logits/rejected": 0.935335636138916,
      "logps/chosen": -119.57994079589844,
      "logps/rejected": -112.05470275878906,
      "loss": 0.0099,
      "rewards/accuracies": 0.3125,
      "rewards/chosen": -0.0007582681137137115,
      "rewards/margins": 0.0003905483172275126,
      "rewards/rejected": -0.001148816430941224,
      "step": 180
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.0,
      "logits/chosen": 0.8246952295303345,
      "logits/rejected": 0.8706514239311218,
      "logps/chosen": -73.35179138183594,
      "logps/rejected": -76.91303253173828,
      "loss": 0.0099,
      "rewards/accuracies": 0.25,
      "rewards/chosen": -0.0005546126631088555,
      "rewards/margins": 0.0004530520236585289,
      "rewards/rejected": -0.0010076647158712149,
      "step": 190
    },
    {
      "epoch": 1.98,
      "step": 190,
      "total_flos": 0.0,
      "train_loss": 0.00995884225458691,
      "train_runtime": 2568.9627,
      "train_samples_per_second": 4.759,
      "train_steps_per_second": 0.074
    }
  ],
  "logging_steps": 10,
  "max_steps": 190,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}