File size: 5,200 Bytes
1e7be6a
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.383378016085791,
  "eval_steps": 500,
  "global_step": 7000,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.6702412868632708,
      "grad_norm": 23.28346824645996,
      "learning_rate": 9.798511296679954e-06,
      "loss": 0.5531,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8025477528572083,
      "eval_loss": 0.4079710841178894,
      "eval_runtime": 9.2612,
      "eval_samples_per_second": 322.097,
      "eval_steps_per_second": 20.192,
      "step": 746
    },
    {
      "epoch": 1.3404825737265416,
      "grad_norm": 38.64032745361328,
      "learning_rate": 9.094595255251797e-06,
      "loss": 0.3957,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8276902437210083,
      "eval_loss": 0.42224231362342834,
      "eval_runtime": 9.5099,
      "eval_samples_per_second": 313.673,
      "eval_steps_per_second": 19.664,
      "step": 1492
    },
    {
      "epoch": 2.0107238605898123,
      "grad_norm": 34.970176696777344,
      "learning_rate": 8.390679213823639e-06,
      "loss": 0.3185,
      "step": 1500
    },
    {
      "epoch": 2.680965147453083,
      "grad_norm": 13.735482215881348,
      "learning_rate": 7.686763172395482e-06,
      "loss": 0.1887,
      "step": 2000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.819309413433075,
      "eval_loss": 0.49816903471946716,
      "eval_runtime": 9.3819,
      "eval_samples_per_second": 317.952,
      "eval_steps_per_second": 19.932,
      "step": 2238
    },
    {
      "epoch": 3.351206434316354,
      "grad_norm": 10.144265174865723,
      "learning_rate": 6.982847130967324e-06,
      "loss": 0.1522,
      "step": 2500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8313778042793274,
      "eval_loss": 0.6922349333763123,
      "eval_runtime": 9.2835,
      "eval_samples_per_second": 321.324,
      "eval_steps_per_second": 20.143,
      "step": 2984
    },
    {
      "epoch": 4.021447721179625,
      "grad_norm": 39.181697845458984,
      "learning_rate": 6.278931089539166e-06,
      "loss": 0.1152,
      "step": 3000
    },
    {
      "epoch": 4.6916890080428955,
      "grad_norm": 5.586261749267578,
      "learning_rate": 5.575015048111009e-06,
      "loss": 0.0669,
      "step": 3500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8253436088562012,
      "eval_loss": 0.8140854835510254,
      "eval_runtime": 9.3552,
      "eval_samples_per_second": 318.859,
      "eval_steps_per_second": 19.989,
      "step": 3730
    },
    {
      "epoch": 5.361930294906166,
      "grad_norm": 0.027025775983929634,
      "learning_rate": 4.871099006682851e-06,
      "loss": 0.0526,
      "step": 4000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8223265409469604,
      "eval_loss": 0.9023452997207642,
      "eval_runtime": 9.2601,
      "eval_samples_per_second": 322.135,
      "eval_steps_per_second": 20.194,
      "step": 4476
    },
    {
      "epoch": 6.032171581769437,
      "grad_norm": 0.05576400086283684,
      "learning_rate": 4.167182965254693e-06,
      "loss": 0.0417,
      "step": 4500
    },
    {
      "epoch": 6.702412868632708,
      "grad_norm": 55.70774841308594,
      "learning_rate": 3.4632669238265355e-06,
      "loss": 0.0301,
      "step": 5000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8266845345497131,
      "eval_loss": 0.9764488339424133,
      "eval_runtime": 9.3516,
      "eval_samples_per_second": 318.981,
      "eval_steps_per_second": 19.996,
      "step": 5222
    },
    {
      "epoch": 7.372654155495979,
      "grad_norm": 0.9753480553627014,
      "learning_rate": 2.759350882398378e-06,
      "loss": 0.0218,
      "step": 5500
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8320482969284058,
      "eval_loss": 0.9748868346214294,
      "eval_runtime": 9.391,
      "eval_samples_per_second": 317.644,
      "eval_steps_per_second": 19.913,
      "step": 5968
    },
    {
      "epoch": 8.04289544235925,
      "grad_norm": 65.11773681640625,
      "learning_rate": 2.0554348409702203e-06,
      "loss": 0.0196,
      "step": 6000
    },
    {
      "epoch": 8.71313672922252,
      "grad_norm": 0.08348937332630157,
      "learning_rate": 1.3515187995420628e-06,
      "loss": 0.0136,
      "step": 6500
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8246731758117676,
      "eval_loss": 1.0319347381591797,
      "eval_runtime": 9.3631,
      "eval_samples_per_second": 318.591,
      "eval_steps_per_second": 19.972,
      "step": 6714
    },
    {
      "epoch": 9.383378016085791,
      "grad_norm": 5.880457401275635,
      "learning_rate": 6.476027581139051e-07,
      "loss": 0.0095,
      "step": 7000
    }
  ],
  "logging_steps": 500,
  "max_steps": 7460,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 1.845415899511027e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.0502427338108112e-05,
    "per_device_train_batch_size": 16
  }
}