File size: 9,337 Bytes
8d6f77f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 100,
  "global_step": 162,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 2.9411764705882356e-07,
      "logits/chosen": -2.640578269958496,
      "logits/rejected": -2.6619861125946045,
      "logps/chosen": -410.817138671875,
      "logps/rejected": -784.9041137695312,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.9411764705882355e-06,
      "logits/chosen": -2.580946683883667,
      "logits/rejected": -2.562483549118042,
      "logps/chosen": -348.2420959472656,
      "logps/rejected": -529.5997924804688,
      "loss": 0.6857,
      "rewards/accuracies": 0.5416666865348816,
      "rewards/chosen": 0.047004420310258865,
      "rewards/margins": 0.018747717142105103,
      "rewards/rejected": 0.028256705030798912,
      "step": 10
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.994720857837211e-06,
      "logits/chosen": -2.563664197921753,
      "logits/rejected": -2.5549397468566895,
      "logps/chosen": -303.5776062011719,
      "logps/rejected": -509.2774353027344,
      "loss": 0.6578,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": 0.23815405368804932,
      "rewards/margins": 0.07588066160678864,
      "rewards/rejected": 0.1622733771800995,
      "step": 20
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.901488388458247e-06,
      "logits/chosen": -2.526702404022217,
      "logits/rejected": -2.5158963203430176,
      "logps/chosen": -335.10333251953125,
      "logps/rejected": -552.9954223632812,
      "loss": 0.6058,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/chosen": 0.2050032615661621,
      "rewards/margins": 0.20939965546131134,
      "rewards/rejected": -0.004396387841552496,
      "step": 30
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.6959649910976165e-06,
      "logits/chosen": -2.532031536102295,
      "logits/rejected": -2.4838037490844727,
      "logps/chosen": -300.42132568359375,
      "logps/rejected": -509.403564453125,
      "loss": 0.5705,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/chosen": 0.23074543476104736,
      "rewards/margins": 0.2837901711463928,
      "rewards/rejected": -0.053044695407152176,
      "step": 40
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.387760711393052e-06,
      "logits/chosen": -2.5283217430114746,
      "logits/rejected": -2.46999192237854,
      "logps/chosen": -327.3263244628906,
      "logps/rejected": -512.6666870117188,
      "loss": 0.5487,
      "rewards/accuracies": 0.800000011920929,
      "rewards/chosen": 0.2256808578968048,
      "rewards/margins": 0.36225491762161255,
      "rewards/rejected": -0.13657405972480774,
      "step": 50
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.991286838919086e-06,
      "logits/chosen": -2.508136034011841,
      "logits/rejected": -2.471802234649658,
      "logps/chosen": -333.85711669921875,
      "logps/rejected": -523.9845581054688,
      "loss": 0.5393,
      "rewards/accuracies": 0.784375011920929,
      "rewards/chosen": 0.1835666298866272,
      "rewards/margins": 0.38937950134277344,
      "rewards/rejected": -0.20581285655498505,
      "step": 60
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.5250820513035403e-06,
      "logits/chosen": -2.492502450942993,
      "logits/rejected": -2.484013080596924,
      "logps/chosen": -314.8898620605469,
      "logps/rejected": -589.8179931640625,
      "loss": 0.5165,
      "rewards/accuracies": 0.84375,
      "rewards/chosen": 0.19387896358966827,
      "rewards/margins": 0.5265167951583862,
      "rewards/rejected": -0.33263787627220154,
      "step": 70
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.0109455662659126e-06,
      "logits/chosen": -2.527844190597534,
      "logits/rejected": -2.502004623413086,
      "logps/chosen": -320.5412292480469,
      "logps/rejected": -579.2618408203125,
      "loss": 0.5083,
      "rewards/accuracies": 0.815625011920929,
      "rewards/chosen": 0.12712647020816803,
      "rewards/margins": 0.5316451787948608,
      "rewards/rejected": -0.4045187532901764,
      "step": 80
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.4729178344249007e-06,
      "logits/chosen": -2.5523924827575684,
      "logits/rejected": -2.5245110988616943,
      "logps/chosen": -341.2242431640625,
      "logps/rejected": -596.2203369140625,
      "loss": 0.496,
      "rewards/accuracies": 0.8125,
      "rewards/chosen": 0.06856798380613327,
      "rewards/margins": 0.590112030506134,
      "rewards/rejected": -0.5215439796447754,
      "step": 90
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.936156434546515e-06,
      "logits/chosen": -2.5596001148223877,
      "logits/rejected": -2.568671703338623,
      "logps/chosen": -333.087158203125,
      "logps/rejected": -611.3602905273438,
      "loss": 0.4768,
      "rewards/accuracies": 0.8187500238418579,
      "rewards/chosen": 0.12124671787023544,
      "rewards/margins": 0.6403753757476807,
      "rewards/rejected": -0.5191286206245422,
      "step": 100
    },
    {
      "epoch": 0.62,
      "eval_logits/chosen": -2.585242509841919,
      "eval_logits/rejected": -2.479508399963379,
      "eval_logps/chosen": -293.3752136230469,
      "eval_logps/rejected": -292.1757507324219,
      "eval_loss": 0.6442223191261292,
      "eval_rewards/accuracies": 0.5659999847412109,
      "eval_rewards/chosen": 0.18966493010520935,
      "eval_rewards/margins": 0.15400375425815582,
      "eval_rewards/rejected": 0.035661179572343826,
      "eval_runtime": 398.0482,
      "eval_samples_per_second": 5.025,
      "eval_steps_per_second": 0.628,
      "step": 100
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.4257597331216211e-06,
      "logits/chosen": -2.5242629051208496,
      "logits/rejected": -2.484691619873047,
      "logps/chosen": -325.24761962890625,
      "logps/rejected": -541.4208984375,
      "loss": 0.4856,
      "rewards/accuracies": 0.8031250238418579,
      "rewards/chosen": 0.11120424419641495,
      "rewards/margins": 0.6161486506462097,
      "rewards/rejected": -0.5049443244934082,
      "step": 110
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.655933126436565e-07,
      "logits/chosen": -2.5429892539978027,
      "logits/rejected": -2.520012855529785,
      "logps/chosen": -329.37347412109375,
      "logps/rejected": -596.545654296875,
      "loss": 0.4791,
      "rewards/accuracies": 0.800000011920929,
      "rewards/chosen": 0.06060720607638359,
      "rewards/margins": 0.6603037714958191,
      "rewards/rejected": -0.5996966361999512,
      "step": 120
    },
    {
      "epoch": 0.8,
      "learning_rate": 5.771740434959278e-07,
      "logits/chosen": -2.5939841270446777,
      "logits/rejected": -2.543733835220337,
      "logps/chosen": -326.51824951171875,
      "logps/rejected": -585.2608032226562,
      "loss": 0.4705,
      "rewards/accuracies": 0.831250011920929,
      "rewards/chosen": 0.0589924156665802,
      "rewards/margins": 0.7284678220748901,
      "rewards/rejected": -0.6694754362106323,
      "step": 130
    },
    {
      "epoch": 0.86,
      "learning_rate": 2.786639790067719e-07,
      "logits/chosen": -2.538792848587036,
      "logits/rejected": -2.572946548461914,
      "logps/chosen": -327.40155029296875,
      "logps/rejected": -593.7877197265625,
      "loss": 0.4828,
      "rewards/accuracies": 0.828125,
      "rewards/chosen": 0.006606454961001873,
      "rewards/margins": 0.6702617406845093,
      "rewards/rejected": -0.6636553406715393,
      "step": 140
    },
    {
      "epoch": 0.93,
      "learning_rate": 8.402111802159413e-08,
      "logits/chosen": -2.60109281539917,
      "logits/rejected": -2.500326633453369,
      "logps/chosen": -350.4434814453125,
      "logps/rejected": -545.3392333984375,
      "loss": 0.4729,
      "rewards/accuracies": 0.800000011920929,
      "rewards/chosen": 0.03271277993917465,
      "rewards/margins": 0.6479501128196716,
      "rewards/rejected": -0.615237295627594,
      "step": 150
    },
    {
      "epoch": 0.99,
      "learning_rate": 2.34674439005822e-09,
      "logits/chosen": -2.6234354972839355,
      "logits/rejected": -2.563753366470337,
      "logps/chosen": -370.4894104003906,
      "logps/rejected": -656.5492553710938,
      "loss": 0.4684,
      "rewards/accuracies": 0.784375011920929,
      "rewards/chosen": 0.009518811479210854,
      "rewards/margins": 0.7224863171577454,
      "rewards/rejected": -0.7129674553871155,
      "step": 160
    },
    {
      "epoch": 1.0,
      "step": 162,
      "total_flos": 0.0,
      "train_loss": 0.5280830058050744,
      "train_runtime": 8024.3189,
      "train_samples_per_second": 2.584,
      "train_steps_per_second": 0.02
    }
  ],
  "logging_steps": 10,
  "max_steps": 162,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}