File size: 9,454 Bytes
389395f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9992429977289932,
  "eval_steps": 500,
  "global_step": 165,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "grad_norm": 20.867678760375714,
      "learning_rate": 2.941176470588235e-08,
      "logits/chosen": -2.0811095237731934,
      "logits/rejected": -1.7104089260101318,
      "logps/chosen": -188.16197204589844,
      "logps/rejected": -218.83871459960938,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.06,
      "grad_norm": 19.74353822187935,
      "learning_rate": 2.941176470588235e-07,
      "logits/chosen": -1.8544065952301025,
      "logits/rejected": -1.8290244340896606,
      "logps/chosen": -191.76025390625,
      "logps/rejected": -211.63970947265625,
      "loss": 0.6925,
      "rewards/accuracies": 0.4583333432674408,
      "rewards/chosen": -0.01668192446231842,
      "rewards/margins": 0.0031901856418699026,
      "rewards/rejected": -0.019872108474373817,
      "step": 10
    },
    {
      "epoch": 0.12,
      "grad_norm": 25.92045039305084,
      "learning_rate": 4.994932636402031e-07,
      "logits/chosen": -1.4885847568511963,
      "logits/rejected": -1.4829342365264893,
      "logps/chosen": -226.84854125976562,
      "logps/rejected": -239.1204833984375,
      "loss": 0.6885,
      "rewards/accuracies": 0.559374988079071,
      "rewards/chosen": -0.26655465364456177,
      "rewards/margins": 0.0009398255497217178,
      "rewards/rejected": -0.2674944996833801,
      "step": 20
    },
    {
      "epoch": 0.18,
      "grad_norm": 41.251364875481954,
      "learning_rate": 4.905416503522123e-07,
      "logits/chosen": -1.1668879985809326,
      "logits/rejected": -1.1640729904174805,
      "logps/chosen": -190.57748413085938,
      "logps/rejected": -212.91455078125,
      "loss": 0.673,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": -0.09980447590351105,
      "rewards/margins": 0.06015855073928833,
      "rewards/rejected": -0.15996304154396057,
      "step": 30
    },
    {
      "epoch": 0.24,
      "grad_norm": 28.11499401185622,
      "learning_rate": 4.707922373336523e-07,
      "logits/chosen": -0.7021564245223999,
      "logits/rejected": -0.6727257370948792,
      "logps/chosen": -210.4853057861328,
      "logps/rejected": -240.263916015625,
      "loss": 0.662,
      "rewards/accuracies": 0.609375,
      "rewards/chosen": -0.24566245079040527,
      "rewards/margins": 0.12348496913909912,
      "rewards/rejected": -0.369147390127182,
      "step": 40
    },
    {
      "epoch": 0.3,
      "grad_norm": 30.137593860562752,
      "learning_rate": 4.4113156629677313e-07,
      "logits/chosen": -0.23419609665870667,
      "logits/rejected": -0.20015835762023926,
      "logps/chosen": -269.1314697265625,
      "logps/rejected": -288.4267578125,
      "loss": 0.6737,
      "rewards/accuracies": 0.578125,
      "rewards/chosen": -0.7066656351089478,
      "rewards/margins": 0.08174248039722443,
      "rewards/rejected": -0.788408100605011,
      "step": 50
    },
    {
      "epoch": 0.36,
      "grad_norm": 26.750008876948538,
      "learning_rate": 4.0289109058972283e-07,
      "logits/chosen": -0.2947733402252197,
      "logits/rejected": -0.2833467125892639,
      "logps/chosen": -213.510498046875,
      "logps/rejected": -245.3372344970703,
      "loss": 0.6631,
      "rewards/accuracies": 0.6468750238418579,
      "rewards/chosen": -0.27190494537353516,
      "rewards/margins": 0.15612241625785828,
      "rewards/rejected": -0.42802733182907104,
      "step": 60
    },
    {
      "epoch": 0.42,
      "grad_norm": 30.782722779567703,
      "learning_rate": 3.577874068920446e-07,
      "logits/chosen": -0.11177041381597519,
      "logits/rejected": -0.028942082077264786,
      "logps/chosen": -228.6294708251953,
      "logps/rejected": -266.2515563964844,
      "loss": 0.6631,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.30979710817337036,
      "rewards/margins": 0.17394262552261353,
      "rewards/rejected": -0.48373979330062866,
      "step": 70
    },
    {
      "epoch": 0.48,
      "grad_norm": 32.927052210750006,
      "learning_rate": 3.078451980100854e-07,
      "logits/chosen": 0.29397937655448914,
      "logits/rejected": 0.5637707710266113,
      "logps/chosen": -242.727294921875,
      "logps/rejected": -275.2801208496094,
      "loss": 0.6552,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.5540092587471008,
      "rewards/margins": 0.17995788156986237,
      "rewards/rejected": -0.733967125415802,
      "step": 80
    },
    {
      "epoch": 0.55,
      "grad_norm": 23.357021774601378,
      "learning_rate": 2.553063458334059e-07,
      "logits/chosen": 0.48601818084716797,
      "logits/rejected": 0.49343404173851013,
      "logps/chosen": -277.7685241699219,
      "logps/rejected": -304.49053955078125,
      "loss": 0.6674,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.9620914459228516,
      "rewards/margins": 0.14792589843273163,
      "rewards/rejected": -1.1100172996520996,
      "step": 90
    },
    {
      "epoch": 0.61,
      "grad_norm": 25.083141130465293,
      "learning_rate": 2.0252929432814287e-07,
      "logits/chosen": -0.11149320751428604,
      "logits/rejected": 0.00033601821633055806,
      "logps/chosen": -241.638427734375,
      "logps/rejected": -270.65484619140625,
      "loss": 0.6521,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.4907761514186859,
      "rewards/margins": 0.14960728585720062,
      "rewards/rejected": -0.6403835415840149,
      "step": 100
    },
    {
      "epoch": 0.67,
      "grad_norm": 24.59487007227396,
      "learning_rate": 1.5188318011445906e-07,
      "logits/chosen": 0.22773680090904236,
      "logits/rejected": 0.27657338976860046,
      "logps/chosen": -212.44287109375,
      "logps/rejected": -249.192138671875,
      "loss": 0.6713,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.32515043020248413,
      "rewards/margins": 0.17877240478992462,
      "rewards/rejected": -0.5039228200912476,
      "step": 110
    },
    {
      "epoch": 0.73,
      "grad_norm": 28.591902538050096,
      "learning_rate": 1.0564148305586295e-07,
      "logits/chosen": 0.18790467083454132,
      "logits/rejected": 0.32668787240982056,
      "logps/chosen": -234.372314453125,
      "logps/rejected": -267.2543029785156,
      "loss": 0.6716,
      "rewards/accuracies": 0.640625,
      "rewards/chosen": -0.5469544529914856,
      "rewards/margins": 0.13245931267738342,
      "rewards/rejected": -0.6794136762619019,
      "step": 120
    },
    {
      "epoch": 0.79,
      "grad_norm": 26.169080834485463,
      "learning_rate": 6.587997083462196e-08,
      "logits/chosen": 0.09704655408859253,
      "logits/rejected": 0.28270775079727173,
      "logps/chosen": -253.8743438720703,
      "logps/rejected": -273.24005126953125,
      "loss": 0.658,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.5650860071182251,
      "rewards/margins": 0.14433643221855164,
      "rewards/rejected": -0.7094224691390991,
      "step": 130
    },
    {
      "epoch": 0.85,
      "grad_norm": 31.142574487385833,
      "learning_rate": 3.438351873250492e-08,
      "logits/chosen": 0.19891077280044556,
      "logits/rejected": 0.20664291083812714,
      "logps/chosen": -233.57754516601562,
      "logps/rejected": -270.820068359375,
      "loss": 0.6486,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.5764339566230774,
      "rewards/margins": 0.19204215705394745,
      "rewards/rejected": -0.7684761881828308,
      "step": 140
    },
    {
      "epoch": 0.91,
      "grad_norm": 26.358654813660795,
      "learning_rate": 1.256598743236703e-08,
      "logits/chosen": 0.22453470528125763,
      "logits/rejected": 0.14322993159294128,
      "logps/chosen": -235.47705078125,
      "logps/rejected": -279.6608581542969,
      "loss": 0.649,
      "rewards/accuracies": 0.6656249761581421,
      "rewards/chosen": -0.5816648006439209,
      "rewards/margins": 0.2078080177307129,
      "rewards/rejected": -0.789472758769989,
      "step": 150
    },
    {
      "epoch": 0.97,
      "grad_norm": 27.022153331528376,
      "learning_rate": 1.406755487774386e-09,
      "logits/chosen": 0.23891225457191467,
      "logits/rejected": 0.44686970114707947,
      "logps/chosen": -250.3038787841797,
      "logps/rejected": -281.60113525390625,
      "loss": 0.6594,
      "rewards/accuracies": 0.6031249761581421,
      "rewards/chosen": -0.6393707394599915,
      "rewards/margins": 0.1696932017803192,
      "rewards/rejected": -0.8090640306472778,
      "step": 160
    },
    {
      "epoch": 1.0,
      "step": 165,
      "total_flos": 0.0,
      "train_loss": 0.6651211370121349,
      "train_runtime": 19266.0029,
      "train_samples_per_second": 1.097,
      "train_steps_per_second": 0.009
    }
  ],
  "logging_steps": 10,
  "max_steps": 165,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 200,
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}