File size: 4,337 Bytes
e55479e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.733504037817609,
  "global_step": 4400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16,
      "learning_rate": 0.0002388,
      "loss": 4.5334,
      "step": 400
    },
    {
      "epoch": 0.16,
      "eval_loss": 2.026416063308716,
      "eval_runtime": 471.8471,
      "eval_samples_per_second": 10.758,
      "eval_steps_per_second": 1.346,
      "eval_wer": 0.9800369528744665,
      "step": 400
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00029881808566895816,
      "loss": 1.0808,
      "step": 800
    },
    {
      "epoch": 0.32,
      "eval_loss": 0.7069017887115479,
      "eval_runtime": 472.368,
      "eval_samples_per_second": 10.746,
      "eval_steps_per_second": 1.344,
      "eval_wer": 0.593412194448574,
      "step": 800
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0002972316234796404,
      "loss": 0.7499,
      "step": 1200
    },
    {
      "epoch": 0.47,
      "eval_loss": 0.5792935490608215,
      "eval_runtime": 471.2762,
      "eval_samples_per_second": 10.771,
      "eval_steps_per_second": 1.347,
      "eval_wer": 0.49153694225582434,
      "step": 1200
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00029564516129032255,
      "loss": 0.6397,
      "step": 1600
    },
    {
      "epoch": 0.63,
      "eval_loss": 0.5181192755699158,
      "eval_runtime": 470.1504,
      "eval_samples_per_second": 10.797,
      "eval_steps_per_second": 1.351,
      "eval_wer": 0.41618705799902306,
      "step": 1600
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0002940586991010047,
      "loss": 0.597,
      "step": 2000
    },
    {
      "epoch": 0.79,
      "eval_loss": 0.47279468178749084,
      "eval_runtime": 470.9283,
      "eval_samples_per_second": 10.779,
      "eval_steps_per_second": 1.348,
      "eval_wer": 0.37248072716461017,
      "step": 2000
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00029247223691168694,
      "loss": 0.5666,
      "step": 2400
    },
    {
      "epoch": 0.95,
      "eval_loss": 0.4448830783367157,
      "eval_runtime": 471.0174,
      "eval_samples_per_second": 10.777,
      "eval_steps_per_second": 1.348,
      "eval_wer": 0.3520292225030263,
      "step": 2400
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0002908857747223691,
      "loss": 0.5067,
      "step": 2800
    },
    {
      "epoch": 1.1,
      "eval_loss": 0.421601802110672,
      "eval_runtime": 470.0998,
      "eval_samples_per_second": 10.798,
      "eval_steps_per_second": 1.351,
      "eval_wer": 0.3330855650179455,
      "step": 2800
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0002892993125330513,
      "loss": 0.477,
      "step": 3200
    },
    {
      "epoch": 1.26,
      "eval_loss": 0.39891988039016724,
      "eval_runtime": 471.7321,
      "eval_samples_per_second": 10.76,
      "eval_steps_per_second": 1.346,
      "eval_wer": 0.30331089260305394,
      "step": 3200
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00028771285034373344,
      "loss": 0.4633,
      "step": 3600
    },
    {
      "epoch": 1.42,
      "eval_loss": 0.41667959094047546,
      "eval_runtime": 473.0088,
      "eval_samples_per_second": 10.731,
      "eval_steps_per_second": 1.342,
      "eval_wer": 0.3046276042219721,
      "step": 3600
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0002861263881544156,
      "loss": 0.4606,
      "step": 4000
    },
    {
      "epoch": 1.58,
      "eval_loss": 0.3892616033554077,
      "eval_runtime": 473.5469,
      "eval_samples_per_second": 10.719,
      "eval_steps_per_second": 1.341,
      "eval_wer": 0.29747063945462654,
      "step": 4000
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0002845399259650978,
      "loss": 0.4385,
      "step": 4400
    },
    {
      "epoch": 1.73,
      "eval_loss": 0.3867528438568115,
      "eval_runtime": 471.6975,
      "eval_samples_per_second": 10.761,
      "eval_steps_per_second": 1.346,
      "eval_wer": 0.3090237220464247,
      "step": 4400
    }
  ],
  "max_steps": 76140,
  "num_train_epochs": 30,
  "total_flos": 5.530258185352643e+18,
  "trial_name": null,
  "trial_params": null
}