File size: 3,615 Bytes
afef1df
6a678d9
afef1df
 
 
 
 
 
 
 
 
 
6a678d9
 
 
 
 
 
 
 
 
afef1df
 
 
 
6a678d9
 
 
 
 
 
 
 
 
afef1df
 
 
 
6a678d9
 
 
 
 
 
 
 
 
afef1df
 
 
 
 
6a678d9
 
 
 
 
 
 
 
 
afef1df
 
 
 
6a678d9
 
 
 
 
 
 
 
 
afef1df
 
 
 
6a678d9
 
 
 
 
 
 
 
 
afef1df
 
 
 
 
6a678d9
 
 
 
 
 
 
 
afef1df
 
 
 
 
 
 
 
 
 
 
 
 
 
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": 0.4733425974845886,
  "best_model_checkpoint": "./zephyr/05-04-24-Weni-WeniGPT-Agents-Zephyr-1.0.1-KTO_testing kto dataset during training-3_max_steps-145_batch_16_2024-04-05_ppid_9/checkpoint-100",
  "epoch": 0.684931506849315,
  "eval_steps": 50,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14,
      "grad_norm": 8.764093399047852,
      "kl": 11.000414848327637,
      "learning_rate": 0.00018285714285714286,
      "logps/chosen": -320.3092041015625,
      "logps/rejected": -296.3908386230469,
      "loss": 0.432,
      "rewards/chosen": 1.4254474639892578,
      "rewards/margins": 1.3692480325698853,
      "rewards/rejected": 0.0951162651181221,
      "step": 20
    },
    {
      "epoch": 0.27,
      "grad_norm": 5.878363609313965,
      "kl": 1.2673273086547852,
      "learning_rate": 0.0001542857142857143,
      "logps/chosen": -307.4684753417969,
      "logps/rejected": -375.4805603027344,
      "loss": 0.4081,
      "rewards/chosen": -1.0870614051818848,
      "rewards/margins": 3.670943021774292,
      "rewards/rejected": -4.587001800537109,
      "step": 40
    },
    {
      "epoch": 0.34,
      "eval_kl": 1.7102612257003784,
      "eval_logps/chosen": -316.2467346191406,
      "eval_logps/rejected": -332.64117431640625,
      "eval_loss": 0.38967660069465637,
      "eval_rewards/chosen": -0.9805464744567871,
      "eval_rewards/margins": 3.1516332626342773,
      "eval_rewards/rejected": -3.9925851821899414,
      "eval_runtime": 138.3084,
      "eval_samples_per_second": 2.169,
      "eval_steps_per_second": 0.542,
      "step": 50
    },
    {
      "epoch": 0.41,
      "grad_norm": 12.671459197998047,
      "kl": 2.424811840057373,
      "learning_rate": 0.00012714285714285714,
      "logps/chosen": -321.6191101074219,
      "logps/rejected": -360.3143310546875,
      "loss": 0.4136,
      "rewards/chosen": -0.7625396847724915,
      "rewards/margins": 2.869812250137329,
      "rewards/rejected": -3.8002796173095703,
      "step": 60
    },
    {
      "epoch": 0.55,
      "grad_norm": 6.753478050231934,
      "kl": 5.016882419586182,
      "learning_rate": 9.857142857142858e-05,
      "logps/chosen": -314.6625671386719,
      "logps/rejected": -389.6596374511719,
      "loss": 0.4034,
      "rewards/chosen": -1.544058084487915,
      "rewards/margins": 5.419744968414307,
      "rewards/rejected": -7.084101676940918,
      "step": 80
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.0027679901104420424,
      "kl": 0.5703033208847046,
      "learning_rate": 7e-05,
      "logps/chosen": -532.3701171875,
      "logps/rejected": -549.9442749023438,
      "loss": 0.4304,
      "rewards/chosen": -21.35959243774414,
      "rewards/margins": 2.940380096435547,
      "rewards/rejected": -22.644678115844727,
      "step": 100
    },
    {
      "epoch": 0.68,
      "eval_kl": 0.0,
      "eval_logps/chosen": -1804.134765625,
      "eval_logps/rejected": -1664.4483642578125,
      "eval_loss": 0.4733425974845886,
      "eval_rewards/chosen": -149.7693634033203,
      "eval_rewards/margins": -9.425224304199219,
      "eval_rewards/rejected": -137.1732940673828,
      "eval_runtime": 138.3606,
      "eval_samples_per_second": 2.168,
      "eval_steps_per_second": 0.542,
      "step": 100
    }
  ],
  "logging_steps": 20,
  "max_steps": 145,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}