File size: 4,321 Bytes
a7fc36e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": 0.7474747474747475,
  "best_model_checkpoint": "distilhubert-finetuned-not-a-word2/run-2/checkpoint-80",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 80,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.31,
      "grad_norm": 0.8302626013755798,
      "learning_rate": 9.517539799539297e-06,
      "loss": 0.6966,
      "step": 5
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.6829280257225037,
      "learning_rate": 1.9035079599078593e-05,
      "loss": 0.6813,
      "step": 10
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5496665835380554,
      "learning_rate": 2.8552619398617893e-05,
      "loss": 0.6451,
      "step": 15
    },
    {
      "epoch": 1.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.6860618591308594,
      "eval_runtime": 1.3401,
      "eval_samples_per_second": 47.759,
      "eval_steps_per_second": 5.97,
      "step": 16
    },
    {
      "epoch": 1.25,
      "grad_norm": 1.3569624423980713,
      "learning_rate": 2.9610123820788925e-05,
      "loss": 0.6113,
      "step": 20
    },
    {
      "epoch": 1.56,
      "grad_norm": 0.3838456869125366,
      "learning_rate": 2.8552619398617893e-05,
      "loss": 0.5997,
      "step": 25
    },
    {
      "epoch": 1.88,
      "grad_norm": 0.46842828392982483,
      "learning_rate": 2.7495114976446858e-05,
      "loss": 0.6133,
      "step": 30
    },
    {
      "epoch": 2.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.719573974609375,
      "eval_runtime": 1.3426,
      "eval_samples_per_second": 47.668,
      "eval_steps_per_second": 5.958,
      "step": 32
    },
    {
      "epoch": 2.19,
      "grad_norm": 0.7610273957252502,
      "learning_rate": 2.6437610554275826e-05,
      "loss": 0.5971,
      "step": 35
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.5527757406234741,
      "learning_rate": 2.5380106132104794e-05,
      "loss": 0.6373,
      "step": 40
    },
    {
      "epoch": 2.81,
      "grad_norm": 0.9699375033378601,
      "learning_rate": 2.4322601709933762e-05,
      "loss": 0.5701,
      "step": 45
    },
    {
      "epoch": 3.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.6987495422363281,
      "eval_runtime": 1.344,
      "eval_samples_per_second": 47.619,
      "eval_steps_per_second": 5.952,
      "step": 48
    },
    {
      "epoch": 3.12,
      "grad_norm": 0.5822675824165344,
      "learning_rate": 2.3265097287762727e-05,
      "loss": 0.5492,
      "step": 50
    },
    {
      "epoch": 3.44,
      "grad_norm": 0.9859365224838257,
      "learning_rate": 2.2207592865591692e-05,
      "loss": 0.5552,
      "step": 55
    },
    {
      "epoch": 3.75,
      "grad_norm": 0.7946156859397888,
      "learning_rate": 2.115008844342066e-05,
      "loss": 0.5459,
      "step": 60
    },
    {
      "epoch": 4.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.6806888580322266,
      "eval_runtime": 1.3463,
      "eval_samples_per_second": 47.538,
      "eval_steps_per_second": 5.942,
      "step": 64
    },
    {
      "epoch": 4.06,
      "grad_norm": 1.2706314325332642,
      "learning_rate": 2.009258402124963e-05,
      "loss": 0.5792,
      "step": 65
    },
    {
      "epoch": 4.38,
      "grad_norm": 1.456335186958313,
      "learning_rate": 1.9035079599078593e-05,
      "loss": 0.4967,
      "step": 70
    },
    {
      "epoch": 4.69,
      "grad_norm": 1.6856915950775146,
      "learning_rate": 1.797757517690756e-05,
      "loss": 0.4966,
      "step": 75
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.080612659454346,
      "learning_rate": 1.692007075473653e-05,
      "loss": 0.4589,
      "step": 80
    },
    {
      "epoch": 5.0,
      "eval_f1": 0.7474747474747475,
      "eval_loss": 0.6495094299316406,
      "eval_runtime": 1.348,
      "eval_samples_per_second": 47.477,
      "eval_steps_per_second": 5.935,
      "step": 80
    }
  ],
  "logging_steps": 5,
  "max_steps": 160,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 4530217300158096.0,
  "train_batch_size": 12,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 3.045612735852575e-05,
    "per_device_train_batch_size": 12
  }
}