File size: 3,367 Bytes
32da4f3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
{
  "best_metric": 0.8255746290369508,
  "best_model_checkpoint": "/home/tmnam/Desktop/crosslingual-mining-for-domain-nli/output/pretraining/vihealthbert-w_mlm-ViMedNLI/lr3e-5_wr0.1_wd0.0/checkpoint-5000",
  "epoch": 63.1578947368421,
  "eval_steps": 1000,
  "global_step": 6000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.010526315789473684,
      "grad_norm": 44.032386779785156,
      "learning_rate": 1e-08,
      "loss": 11.5293,
      "step": 1
    },
    {
      "epoch": 10.526315789473685,
      "grad_norm": 5.0217604637146,
      "learning_rate": 9.999999999999999e-06,
      "loss": 5.5327,
      "step": 1000
    },
    {
      "epoch": 10.526315789473685,
      "eval_accuracy": 0.588971337116252,
      "eval_loss": 2.7528159618377686,
      "eval_runtime": 0.9763,
      "eval_samples_per_second": 408.668,
      "eval_steps_per_second": 13.315,
      "step": 1000
    },
    {
      "epoch": 21.05263157894737,
      "grad_norm": 3.756152629852295,
      "learning_rate": 1.9999999999999998e-05,
      "loss": 1.9051,
      "step": 2000
    },
    {
      "epoch": 21.05263157894737,
      "eval_accuracy": 0.7783207463349623,
      "eval_loss": 1.467841625213623,
      "eval_runtime": 0.9735,
      "eval_samples_per_second": 409.849,
      "eval_steps_per_second": 13.353,
      "step": 2000
    },
    {
      "epoch": 31.57894736842105,
      "grad_norm": 2.861485481262207,
      "learning_rate": 3e-05,
      "loss": 1.1194,
      "step": 3000
    },
    {
      "epoch": 31.57894736842105,
      "eval_accuracy": 0.8019903409922435,
      "eval_loss": 1.1543285846710205,
      "eval_runtime": 0.9729,
      "eval_samples_per_second": 410.113,
      "eval_steps_per_second": 13.362,
      "step": 3000
    },
    {
      "epoch": 42.10526315789474,
      "grad_norm": 3.267728805541992,
      "learning_rate": 2.8888888888888888e-05,
      "loss": 0.831,
      "step": 4000
    },
    {
      "epoch": 42.10526315789474,
      "eval_accuracy": 0.8146666666666667,
      "eval_loss": 1.097205638885498,
      "eval_runtime": 0.973,
      "eval_samples_per_second": 410.092,
      "eval_steps_per_second": 13.361,
      "step": 4000
    },
    {
      "epoch": 52.63157894736842,
      "grad_norm": 2.2974698543548584,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.6805,
      "step": 5000
    },
    {
      "epoch": 52.63157894736842,
      "eval_accuracy": 0.8255746290369508,
      "eval_loss": 0.9968159794807434,
      "eval_runtime": 0.9748,
      "eval_samples_per_second": 409.295,
      "eval_steps_per_second": 13.335,
      "step": 5000
    },
    {
      "epoch": 63.1578947368421,
      "grad_norm": 2.481663703918457,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 0.5937,
      "step": 6000
    },
    {
      "epoch": 63.1578947368421,
      "eval_accuracy": 0.8242695979172693,
      "eval_loss": 1.0310105085372925,
      "eval_runtime": 0.9734,
      "eval_samples_per_second": 409.886,
      "eval_steps_per_second": 13.355,
      "step": 6000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 30000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 316,
  "save_steps": 1000,
  "total_flos": 1.2522649015636992e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}