File size: 3,728 Bytes
ee17d2b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
{
  "best_metric": 2.0858356952667236,
  "best_model_checkpoint": "./models/checkpoint-70000",
  "epoch": 0.9752155923041558,
  "global_step": 70000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14,
      "learning_rate": 4.7678058113561536e-05,
      "loss": 2.5513,
      "step": 10000
    },
    {
      "epoch": 0.14,
      "eval_gen_len": 19.0,
      "eval_loss": 2.201458692550659,
      "eval_rouge1": 12.1725,
      "eval_rouge2": 2.4216,
      "eval_rougeL": 12.077,
      "eval_rougeLsum": 12.1411,
      "eval_runtime": 1560.686,
      "eval_samples_per_second": 8.565,
      "eval_steps_per_second": 2.141,
      "step": 10000
    },
    {
      "epoch": 0.28,
      "learning_rate": 4.535611622712307e-05,
      "loss": 2.4881,
      "step": 20000
    },
    {
      "epoch": 0.28,
      "eval_gen_len": 19.0,
      "eval_loss": 2.1785953044891357,
      "eval_rouge1": 12.7586,
      "eval_rouge2": 2.6851,
      "eval_rougeL": 12.6518,
      "eval_rougeLsum": 12.733,
      "eval_runtime": 1548.5251,
      "eval_samples_per_second": 8.633,
      "eval_steps_per_second": 2.158,
      "step": 20000
    },
    {
      "epoch": 0.42,
      "learning_rate": 4.3034174340684604e-05,
      "loss": 2.459,
      "step": 30000
    },
    {
      "epoch": 0.42,
      "eval_gen_len": 19.0,
      "eval_loss": 2.1513352394104004,
      "eval_rouge1": 12.7004,
      "eval_rouge2": 2.6318,
      "eval_rougeL": 12.5832,
      "eval_rougeLsum": 12.668,
      "eval_runtime": 1564.1475,
      "eval_samples_per_second": 8.547,
      "eval_steps_per_second": 2.137,
      "step": 30000
    },
    {
      "epoch": 0.56,
      "learning_rate": 4.071223245424614e-05,
      "loss": 2.4361,
      "step": 40000
    },
    {
      "epoch": 0.56,
      "eval_gen_len": 19.0,
      "eval_loss": 2.127916097640991,
      "eval_rouge1": 13.1946,
      "eval_rouge2": 2.8216,
      "eval_rougeL": 13.0738,
      "eval_rougeLsum": 13.1592,
      "eval_runtime": 1584.478,
      "eval_samples_per_second": 8.437,
      "eval_steps_per_second": 2.109,
      "step": 40000
    },
    {
      "epoch": 0.7,
      "learning_rate": 3.8390290567807665e-05,
      "loss": 2.4613,
      "step": 50000
    },
    {
      "epoch": 0.7,
      "eval_gen_len": 19.0,
      "eval_loss": 2.098001003265381,
      "eval_rouge1": 12.8818,
      "eval_rouge2": 2.8541,
      "eval_rougeL": 12.7484,
      "eval_rougeLsum": 12.8289,
      "eval_runtime": 1565.1942,
      "eval_samples_per_second": 8.541,
      "eval_steps_per_second": 2.135,
      "step": 50000
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.60683486813692e-05,
      "loss": 2.462,
      "step": 60000
    },
    {
      "epoch": 0.84,
      "eval_gen_len": 19.0,
      "eval_loss": 2.0897421836853027,
      "eval_rouge1": 13.1141,
      "eval_rouge2": 2.8014,
      "eval_rougeL": 12.9971,
      "eval_rougeLsum": 13.0729,
      "eval_runtime": 1561.4037,
      "eval_samples_per_second": 8.562,
      "eval_steps_per_second": 2.14,
      "step": 60000
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.374640679493074e-05,
      "loss": 2.4478,
      "step": 70000
    },
    {
      "epoch": 0.98,
      "eval_gen_len": 19.0,
      "eval_loss": 2.0858356952667236,
      "eval_rouge1": 13.4344,
      "eval_rouge2": 2.8499,
      "eval_rougeL": 13.3259,
      "eval_rougeLsum": 13.3837,
      "eval_runtime": 1559.9208,
      "eval_samples_per_second": 8.57,
      "eval_steps_per_second": 2.142,
      "step": 70000
    }
  ],
  "max_steps": 215337,
  "num_train_epochs": 3,
  "total_flos": 1.4804990931554304e+17,
  "trial_name": null,
  "trial_params": null
}