Hugging Face's logo

YYYYYYibo
/
nash_dpo_rank4_iter_plus_3

PEFT
Safetensors
mistral
alignment-handbook
Generated from Trainer
trl
dpo
Model card Files Community
nash_dpo_rank4_iter_plus_3
File size: 2,521 Bytes 
d6acae0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
 
{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984,
  "eval_steps": 100,
  "global_step": 39,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 1.25e-06,
      "logits/chosen": -2.5433855056762695,
      "logits/rejected": -2.521193504333496,
      "logps/chosen": -335.29010009765625,
      "logps/rejected": -420.1910400390625,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.646121984004666e-06,
      "logits/chosen": -2.4242465496063232,
      "logits/rejected": -2.3739523887634277,
      "logps/chosen": -283.9295349121094,
      "logps/rejected": -356.6905517578125,
      "loss": 0.6732,
      "rewards/accuracies": 0.6354166865348816,
      "rewards/chosen": 0.0827573761343956,
      "rewards/margins": 0.037400420755147934,
      "rewards/rejected": 0.045356955379247665,
      "step": 10
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.835583164544139e-06,
      "logits/chosen": -2.1755757331848145,
      "logits/rejected": -2.069685935974121,
      "logps/chosen": -307.59771728515625,
      "logps/rejected": -366.8892517089844,
      "loss": 0.6079,
      "rewards/accuracies": 0.768750011920929,
      "rewards/chosen": 0.02839103899896145,
      "rewards/margins": 0.22918327152729034,
      "rewards/rejected": -0.20079222321510315,
      "step": 20
    },
    {
      "epoch": 0.77,
      "learning_rate": 7.723433775328385e-07,
      "logits/chosen": -1.9671818017959595,
      "logits/rejected": -1.8299922943115234,
      "logps/chosen": -337.34051513671875,
      "logps/rejected": -414.0975646972656,
      "loss": 0.5677,
      "rewards/accuracies": 0.746874988079071,
      "rewards/chosen": -0.01181054301559925,
      "rewards/margins": 0.3342409133911133,
      "rewards/rejected": -0.34605151414871216,
      "step": 30
    },
    {
      "epoch": 1.0,
      "step": 39,
      "total_flos": 0.0,
      "train_loss": 0.0,
      "train_runtime": 0.0096,
      "train_samples_per_second": 521173.985,
      "train_steps_per_second": 4065.157
    }
  ],
  "logging_steps": 10,
  "max_steps": 39,
  "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
}