File size: 6,196 Bytes
f8e3a1b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9947089947089947,
  "eval_steps": 1000,
  "global_step": 94,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.010582010582010581,
      "grad_norm": 3.9034635996758364,
      "learning_rate": 5e-08,
      "logits/chosen": -2.8740313053131104,
      "logits/rejected": -2.909637928009033,
      "logps/chosen": -495.3936462402344,
      "logps/rejected": -468.7409973144531,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.10582010582010581,
      "grad_norm": 3.740412497210408,
      "learning_rate": 5e-07,
      "logits/chosen": -2.8803439140319824,
      "logits/rejected": -2.933382987976074,
      "logps/chosen": -489.9436340332031,
      "logps/rejected": -471.76068115234375,
      "loss": 0.6926,
      "rewards/accuracies": 0.4765625,
      "rewards/chosen": 0.0005175346159376204,
      "rewards/margins": 0.001013587461784482,
      "rewards/rejected": -0.0004960527876392007,
      "step": 10
    },
    {
      "epoch": 0.21164021164021163,
      "grad_norm": 4.0026509714343526,
      "learning_rate": 4.82718437161051e-07,
      "logits/chosen": -2.9089906215667725,
      "logits/rejected": -2.9724087715148926,
      "logps/chosen": -490.58831787109375,
      "logps/rejected": -484.2608947753906,
      "loss": 0.677,
      "rewards/accuracies": 0.8101562261581421,
      "rewards/chosen": 0.017761804163455963,
      "rewards/margins": 0.033221058547496796,
      "rewards/rejected": -0.015459256246685982,
      "step": 20
    },
    {
      "epoch": 0.31746031746031744,
      "grad_norm": 5.045224654085605,
      "learning_rate": 4.332629679574565e-07,
      "logits/chosen": -2.9424614906311035,
      "logits/rejected": -2.994748830795288,
      "logps/chosen": -486.4335021972656,
      "logps/rejected": -484.3814392089844,
      "loss": 0.623,
      "rewards/accuracies": 0.8492187261581421,
      "rewards/chosen": 0.045325733721256256,
      "rewards/margins": 0.14735476672649384,
      "rewards/rejected": -0.10202904045581818,
      "step": 30
    },
    {
      "epoch": 0.42328042328042326,
      "grad_norm": 3.94438220021588,
      "learning_rate": 3.584709347793895e-07,
      "logits/chosen": -2.92952036857605,
      "logits/rejected": -3.001981019973755,
      "logps/chosen": -487.70849609375,
      "logps/rejected": -510.1280212402344,
      "loss": 0.5749,
      "rewards/accuracies": 0.85546875,
      "rewards/chosen": 0.017966564744710922,
      "rewards/margins": 0.2901422381401062,
      "rewards/rejected": -0.2721756398677826,
      "step": 40
    },
    {
      "epoch": 0.5291005291005291,
      "grad_norm": 3.0093221435481667,
      "learning_rate": 2.6868252339660607e-07,
      "logits/chosen": -2.92526912689209,
      "logits/rejected": -2.995861768722534,
      "logps/chosen": -514.0173950195312,
      "logps/rejected": -578.7903442382812,
      "loss": 0.4668,
      "rewards/accuracies": 0.8609374761581421,
      "rewards/chosen": -0.239375501871109,
      "rewards/margins": 0.7664871215820312,
      "rewards/rejected": -1.0058627128601074,
      "step": 50
    },
    {
      "epoch": 0.6349206349206349,
      "grad_norm": 3.4147008885194587,
      "learning_rate": 1.763112063972739e-07,
      "logits/chosen": -2.9192681312561035,
      "logits/rejected": -2.9830739498138428,
      "logps/chosen": -526.348876953125,
      "logps/rejected": -614.0721435546875,
      "loss": 0.421,
      "rewards/accuracies": 0.858593761920929,
      "rewards/chosen": -0.4103693962097168,
      "rewards/margins": 0.9783406257629395,
      "rewards/rejected": -1.3887102603912354,
      "step": 60
    },
    {
      "epoch": 0.7407407407407407,
      "grad_norm": 3.0922082498626846,
      "learning_rate": 9.412754953531663e-08,
      "logits/chosen": -2.918205976486206,
      "logits/rejected": -2.970673084259033,
      "logps/chosen": -554.398681640625,
      "logps/rejected": -667.9301147460938,
      "loss": 0.4035,
      "rewards/accuracies": 0.8492187261581421,
      "rewards/chosen": -0.6199524998664856,
      "rewards/margins": 1.1833505630493164,
      "rewards/rejected": -1.8033031225204468,
      "step": 70
    },
    {
      "epoch": 0.8465608465608465,
      "grad_norm": 3.5694688653613595,
      "learning_rate": 3.349364905389032e-08,
      "logits/chosen": -2.8878700733184814,
      "logits/rejected": -2.932638645172119,
      "logps/chosen": -557.0501708984375,
      "logps/rejected": -678.7019653320312,
      "loss": 0.3889,
      "rewards/accuracies": 0.8578125238418579,
      "rewards/chosen": -0.7133009433746338,
      "rewards/margins": 1.2904046773910522,
      "rewards/rejected": -2.0037055015563965,
      "step": 80
    },
    {
      "epoch": 0.9523809523809523,
      "grad_norm": 3.149374311366939,
      "learning_rate": 2.7922934437178692e-09,
      "logits/chosen": -2.89034366607666,
      "logits/rejected": -2.936624765396118,
      "logps/chosen": -562.55322265625,
      "logps/rejected": -680.3013305664062,
      "loss": 0.3864,
      "rewards/accuracies": 0.836718738079071,
      "rewards/chosen": -0.7662609815597534,
      "rewards/margins": 1.2681959867477417,
      "rewards/rejected": -2.034456968307495,
      "step": 90
    },
    {
      "epoch": 0.9947089947089947,
      "step": 94,
      "total_flos": 0.0,
      "train_loss": 0.5087684798747936,
      "train_runtime": 2521.2346,
      "train_samples_per_second": 38.335,
      "train_steps_per_second": 0.037
    }
  ],
  "logging_steps": 10,
  "max_steps": 94,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}