File size: 3,981 Bytes
4c3f658
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
183
184
185
186
187
{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1176,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.13,
      "learning_rate": 1.914965986394558e-05,
      "loss": 0.299,
      "step": 50
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8299319727891158e-05,
      "loss": 0.0765,
      "step": 100
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7448979591836738e-05,
      "loss": 0.0581,
      "step": 150
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6598639455782314e-05,
      "loss": 0.0284,
      "step": 200
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5748299319727894e-05,
      "loss": 0.0415,
      "step": 250
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4897959183673472e-05,
      "loss": 0.0061,
      "step": 300
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4047619047619048e-05,
      "loss": 0.015,
      "step": 350
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3197278911564626e-05,
      "loss": 0.0231,
      "step": 400
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2346938775510204e-05,
      "loss": 0.0154,
      "step": 450
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1496598639455783e-05,
      "loss": 0.0001,
      "step": 500
    },
    {
      "epoch": 1.28,
      "eval_accuracy": 0.9990436722983742,
      "eval_loss": 0.004529774188995361,
      "eval_runtime": 16.1182,
      "eval_samples_per_second": 194.625,
      "eval_steps_per_second": 12.222,
      "step": 500
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.064625850340136e-05,
      "loss": 0.0037,
      "step": 550
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.795918367346939e-06,
      "loss": 0.0082,
      "step": 600
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.945578231292518e-06,
      "loss": 0.0062,
      "step": 650
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.095238095238097e-06,
      "loss": 0.0054,
      "step": 700
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.244897959183675e-06,
      "loss": 0.0175,
      "step": 750
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.394557823129253e-06,
      "loss": 0.0036,
      "step": 800
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.54421768707483e-06,
      "loss": 0.0,
      "step": 850
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.693877551020409e-06,
      "loss": 0.0033,
      "step": 900
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.843537414965986e-06,
      "loss": 0.0056,
      "step": 950
    },
    {
      "epoch": 2.55,
      "learning_rate": 2.993197278911565e-06,
      "loss": 0.0061,
      "step": 1000
    },
    {
      "epoch": 2.55,
      "eval_accuracy": 0.9990436722983742,
      "eval_loss": 0.006005433388054371,
      "eval_runtime": 16.1574,
      "eval_samples_per_second": 194.152,
      "eval_steps_per_second": 12.193,
      "step": 1000
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1428571428571427e-06,
      "loss": 0.0053,
      "step": 1050
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2925170068027212e-06,
      "loss": 0.0022,
      "step": 1100
    },
    {
      "epoch": 2.93,
      "learning_rate": 4.421768707482994e-07,
      "loss": 0.0041,
      "step": 1150
    },
    {
      "epoch": 3.0,
      "step": 1176,
      "total_flos": 7504428268195152.0,
      "train_loss": 0.026978637829607473,
      "train_runtime": 755.5675,
      "train_samples_per_second": 49.735,
      "train_steps_per_second": 1.556
    }
  ],
  "logging_steps": 50,
  "max_steps": 1176,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 7504428268195152.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}