File size: 5,469 Bytes
02160f9
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
{
  "best_metric": 0.7326732673267327,
  "best_model_checkpoint": "distilhubert-finetuned-not-a-word2/run-3/checkpoint-8",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 80,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.62,
      "grad_norm": 0.94398033618927,
      "learning_rate": 7.138113611405621e-06,
      "loss": 0.6981,
      "step": 5
    },
    {
      "epoch": 1.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.6864166259765625,
      "eval_runtime": 1.3249,
      "eval_samples_per_second": 48.305,
      "eval_steps_per_second": 6.038,
      "step": 8
    },
    {
      "epoch": 1.25,
      "grad_norm": 1.5316386222839355,
      "learning_rate": 1.1103732284408743e-05,
      "loss": 0.683,
      "step": 10
    },
    {
      "epoch": 1.88,
      "grad_norm": 0.5028849840164185,
      "learning_rate": 1.031060854980812e-05,
      "loss": 0.6564,
      "step": 15
    },
    {
      "epoch": 2.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.680694580078125,
      "eval_runtime": 1.3565,
      "eval_samples_per_second": 47.179,
      "eval_steps_per_second": 5.897,
      "step": 16
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.5775301456451416,
      "learning_rate": 9.517484815207495e-06,
      "loss": 0.6436,
      "step": 20
    },
    {
      "epoch": 3.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.6852684020996094,
      "eval_runtime": 1.3445,
      "eval_samples_per_second": 47.601,
      "eval_steps_per_second": 5.95,
      "step": 24
    },
    {
      "epoch": 3.12,
      "grad_norm": 0.4549338221549988,
      "learning_rate": 8.72436108060687e-06,
      "loss": 0.6098,
      "step": 25
    },
    {
      "epoch": 3.75,
      "grad_norm": 0.3741567134857178,
      "learning_rate": 7.931237346006246e-06,
      "loss": 0.609,
      "step": 30
    },
    {
      "epoch": 4.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.6952018737792969,
      "eval_runtime": 1.3401,
      "eval_samples_per_second": 47.757,
      "eval_steps_per_second": 5.97,
      "step": 32
    },
    {
      "epoch": 4.38,
      "grad_norm": 0.38969138264656067,
      "learning_rate": 7.138113611405621e-06,
      "loss": 0.6091,
      "step": 35
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.2588692009449005,
      "learning_rate": 6.344989876804997e-06,
      "loss": 0.6043,
      "step": 40
    },
    {
      "epoch": 5.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.7042198181152344,
      "eval_runtime": 1.3815,
      "eval_samples_per_second": 46.326,
      "eval_steps_per_second": 5.791,
      "step": 40
    },
    {
      "epoch": 5.62,
      "grad_norm": 0.6705239415168762,
      "learning_rate": 5.551866142204372e-06,
      "loss": 0.6093,
      "step": 45
    },
    {
      "epoch": 6.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.7079658508300781,
      "eval_runtime": 1.3745,
      "eval_samples_per_second": 46.561,
      "eval_steps_per_second": 5.82,
      "step": 48
    },
    {
      "epoch": 6.25,
      "grad_norm": 0.4161934554576874,
      "learning_rate": 4.758742407603747e-06,
      "loss": 0.5737,
      "step": 50
    },
    {
      "epoch": 6.88,
      "grad_norm": 0.3530200719833374,
      "learning_rate": 3.965618673003123e-06,
      "loss": 0.6355,
      "step": 55
    },
    {
      "epoch": 7.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.7088470458984375,
      "eval_runtime": 1.3736,
      "eval_samples_per_second": 46.592,
      "eval_steps_per_second": 5.824,
      "step": 56
    },
    {
      "epoch": 7.5,
      "grad_norm": 0.7337541580200195,
      "learning_rate": 3.1724949384024984e-06,
      "loss": 0.5738,
      "step": 60
    },
    {
      "epoch": 8.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.7090225219726562,
      "eval_runtime": 1.3562,
      "eval_samples_per_second": 47.192,
      "eval_steps_per_second": 5.899,
      "step": 64
    },
    {
      "epoch": 8.12,
      "grad_norm": 0.3913467526435852,
      "learning_rate": 2.3793712038018737e-06,
      "loss": 0.5872,
      "step": 65
    },
    {
      "epoch": 8.75,
      "grad_norm": 0.6211589574813843,
      "learning_rate": 1.5862474692012492e-06,
      "loss": 0.5549,
      "step": 70
    },
    {
      "epoch": 9.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.7093772888183594,
      "eval_runtime": 1.3631,
      "eval_samples_per_second": 46.952,
      "eval_steps_per_second": 5.869,
      "step": 72
    },
    {
      "epoch": 9.38,
      "grad_norm": 0.30739858746528625,
      "learning_rate": 7.931237346006246e-07,
      "loss": 0.6143,
      "step": 75
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.41105917096138,
      "learning_rate": 0.0,
      "loss": 0.6183,
      "step": 80
    },
    {
      "epoch": 10.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.7090110778808594,
      "eval_runtime": 1.366,
      "eval_samples_per_second": 46.852,
      "eval_steps_per_second": 5.857,
      "step": 80
    }
  ],
  "logging_steps": 5,
  "max_steps": 80,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 1.0052017938770688e+16,
  "train_batch_size": 24,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.1420981778248994e-05,
    "per_device_train_batch_size": 24
  }
}