File size: 3,259 Bytes
d9dcb4b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
{
  "best_metric": 0.027495555579662323,
  "best_model_checkpoint": "./Zephyr/27-03-24-Weni-kto-test_WeniGPT Experiment using KTO trainer with no collator-2_max_steps-786_batch_16_2024-03-27_ppid_1885/checkpoint-100",
  "epoch": 0.7619047619047619,
  "eval_steps": 50,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15,
      "grad_norm": 2.2224462032318115,
      "kl": 0.054589949548244476,
      "learning_rate": 0.0001666666666666667,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 704.2891,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 20
    },
    {
      "epoch": 0.3,
      "grad_norm": 2.9781014919281006,
      "kl": 0.014553820714354515,
      "learning_rate": 0.00019580052493438322,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 371.0873,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 40
    },
    {
      "epoch": 0.38,
      "eval_kl": 0.014955863356590271,
      "eval_logps/chosen": -139.6813507080078,
      "eval_logps/rejected": -288.11480712890625,
      "eval_loss": 0.0440419502556324,
      "eval_rewards/chosen": 4.680752754211426,
      "eval_rewards/margins": 14.199661254882812,
      "eval_rewards/rejected": -9.518909454345703,
      "eval_runtime": 437.1874,
      "eval_samples_per_second": 1.07,
      "eval_steps_per_second": 0.268,
      "step": 50
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.7002557516098022,
      "kl": 0.36391472816467285,
      "learning_rate": 0.0001905511811023622,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 165.1605,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 60
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5644651651382446,
      "kl": 0.01063810009509325,
      "learning_rate": 0.00018556430446194227,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 89.197,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 80
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.18941974639892578,
      "kl": 0.0,
      "learning_rate": 0.0001805774278215223,
      "logps/chosen": NaN,
      "logps/rejected": NaN,
      "loss": 57.9834,
      "rewards/chosen": NaN,
      "rewards/margins": NaN,
      "rewards/rejected": NaN,
      "step": 100
    },
    {
      "epoch": 0.76,
      "eval_kl": 0.0,
      "eval_logps/chosen": -135.0946807861328,
      "eval_logps/rejected": -511.870361328125,
      "eval_loss": 0.027495555579662323,
      "eval_rewards/chosen": 5.139419078826904,
      "eval_rewards/margins": 37.033878326416016,
      "eval_rewards/rejected": -31.89446258544922,
      "eval_runtime": 437.2677,
      "eval_samples_per_second": 1.07,
      "eval_steps_per_second": 0.268,
      "step": 100
    }
  ],
  "logging_steps": 20,
  "max_steps": 786,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}