Mariia commited on
Commit
a564cfc
1 Parent(s): 4e96ba6

Upload trainer_state.json

Browse files
Files changed (1) hide show
  1. trainer_state.json +259 -0
trainer_state.json ADDED
@@ -0,0 +1,259 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9974986599964266,
3
+ "best_model_checkpoint": "./DISTEMIST-huggingface-med/run-4/checkpoint-10632",
4
+ "epoch": 9.53542600896861,
5
+ "global_step": 10632,
6
+ "is_hyper_param_search": true,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.45,
12
+ "learning_rate": 3.718581391614352e-05,
13
+ "loss": 0.1783,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.79,
18
+ "eval_f1": 0.7232514486248266,
19
+ "eval_loss": 0.06977316737174988,
20
+ "eval_runtime": 17.2132,
21
+ "eval_samples_per_second": 517.974,
22
+ "eval_steps_per_second": 64.776,
23
+ "step": 886
24
+ },
25
+ {
26
+ "epoch": 0.9,
27
+ "learning_rate": 3.544000105623068e-05,
28
+ "loss": 0.1046,
29
+ "step": 1000
30
+ },
31
+ {
32
+ "epoch": 1.35,
33
+ "learning_rate": 3.3694188196317836e-05,
34
+ "loss": 0.0712,
35
+ "step": 1500
36
+ },
37
+ {
38
+ "epoch": 1.59,
39
+ "eval_f1": 0.8131887201735358,
40
+ "eval_loss": 0.039787087589502335,
41
+ "eval_runtime": 17.176,
42
+ "eval_samples_per_second": 519.097,
43
+ "eval_steps_per_second": 64.916,
44
+ "step": 1772
45
+ },
46
+ {
47
+ "epoch": 1.79,
48
+ "learning_rate": 3.1948375336404995e-05,
49
+ "loss": 0.0586,
50
+ "step": 2000
51
+ },
52
+ {
53
+ "epoch": 2.24,
54
+ "learning_rate": 3.0202562476492158e-05,
55
+ "loss": 0.0456,
56
+ "step": 2500
57
+ },
58
+ {
59
+ "epoch": 2.38,
60
+ "eval_f1": 0.8951260803706673,
61
+ "eval_loss": 0.020952457562088966,
62
+ "eval_runtime": 17.1751,
63
+ "eval_samples_per_second": 519.123,
64
+ "eval_steps_per_second": 64.919,
65
+ "step": 2658
66
+ },
67
+ {
68
+ "epoch": 2.69,
69
+ "learning_rate": 2.8456749616579317e-05,
70
+ "loss": 0.0318,
71
+ "step": 3000
72
+ },
73
+ {
74
+ "epoch": 3.14,
75
+ "learning_rate": 2.6710936756666475e-05,
76
+ "loss": 0.0277,
77
+ "step": 3500
78
+ },
79
+ {
80
+ "epoch": 3.18,
81
+ "eval_f1": 0.9308534306309513,
82
+ "eval_loss": 0.011393814347684383,
83
+ "eval_runtime": 17.3437,
84
+ "eval_samples_per_second": 514.078,
85
+ "eval_steps_per_second": 64.289,
86
+ "step": 3544
87
+ },
88
+ {
89
+ "epoch": 3.59,
90
+ "learning_rate": 2.496512389675363e-05,
91
+ "loss": 0.0192,
92
+ "step": 4000
93
+ },
94
+ {
95
+ "epoch": 3.97,
96
+ "eval_f1": 0.9564366998577525,
97
+ "eval_loss": 0.008351404219865799,
98
+ "eval_runtime": 17.1902,
99
+ "eval_samples_per_second": 518.667,
100
+ "eval_steps_per_second": 64.862,
101
+ "step": 4430
102
+ },
103
+ {
104
+ "epoch": 4.04,
105
+ "learning_rate": 2.321931103684079e-05,
106
+ "loss": 0.0158,
107
+ "step": 4500
108
+ },
109
+ {
110
+ "epoch": 4.48,
111
+ "learning_rate": 2.147349817692795e-05,
112
+ "loss": 0.0133,
113
+ "step": 5000
114
+ },
115
+ {
116
+ "epoch": 4.77,
117
+ "eval_f1": 0.9696861943283848,
118
+ "eval_loss": 0.00492077087983489,
119
+ "eval_runtime": 17.1702,
120
+ "eval_samples_per_second": 519.271,
121
+ "eval_steps_per_second": 64.938,
122
+ "step": 5316
123
+ },
124
+ {
125
+ "epoch": 4.93,
126
+ "learning_rate": 1.9727685317015108e-05,
127
+ "loss": 0.0108,
128
+ "step": 5500
129
+ },
130
+ {
131
+ "epoch": 5.38,
132
+ "learning_rate": 1.7981872457102267e-05,
133
+ "loss": 0.0079,
134
+ "step": 6000
135
+ },
136
+ {
137
+ "epoch": 5.56,
138
+ "eval_f1": 0.9758359340169416,
139
+ "eval_loss": 0.004112700931727886,
140
+ "eval_runtime": 17.19,
141
+ "eval_samples_per_second": 518.675,
142
+ "eval_steps_per_second": 64.863,
143
+ "step": 6202
144
+ },
145
+ {
146
+ "epoch": 5.83,
147
+ "learning_rate": 1.6236059597189426e-05,
148
+ "loss": 0.0065,
149
+ "step": 6500
150
+ },
151
+ {
152
+ "epoch": 6.28,
153
+ "learning_rate": 1.4490246737276583e-05,
154
+ "loss": 0.0052,
155
+ "step": 7000
156
+ },
157
+ {
158
+ "epoch": 6.36,
159
+ "eval_f1": 0.9876058849754793,
160
+ "eval_loss": 0.0018256960902363062,
161
+ "eval_runtime": 17.1652,
162
+ "eval_samples_per_second": 519.424,
163
+ "eval_steps_per_second": 64.957,
164
+ "step": 7088
165
+ },
166
+ {
167
+ "epoch": 6.73,
168
+ "learning_rate": 1.2744433877363742e-05,
169
+ "loss": 0.0046,
170
+ "step": 7500
171
+ },
172
+ {
173
+ "epoch": 7.15,
174
+ "eval_f1": 0.9915080003575579,
175
+ "eval_loss": 0.001805759035050869,
176
+ "eval_runtime": 17.2049,
177
+ "eval_samples_per_second": 518.225,
178
+ "eval_steps_per_second": 64.807,
179
+ "step": 7974
180
+ },
181
+ {
182
+ "epoch": 7.17,
183
+ "learning_rate": 1.0998621017450902e-05,
184
+ "loss": 0.0032,
185
+ "step": 8000
186
+ },
187
+ {
188
+ "epoch": 7.62,
189
+ "learning_rate": 9.25280815753806e-06,
190
+ "loss": 0.0034,
191
+ "step": 8500
192
+ },
193
+ {
194
+ "epoch": 7.95,
195
+ "eval_f1": 0.9946399857066286,
196
+ "eval_loss": 0.0007319775759242475,
197
+ "eval_runtime": 17.3742,
198
+ "eval_samples_per_second": 513.174,
199
+ "eval_steps_per_second": 64.176,
200
+ "step": 8860
201
+ },
202
+ {
203
+ "epoch": 8.07,
204
+ "learning_rate": 7.5069952976252175e-06,
205
+ "loss": 0.0026,
206
+ "step": 9000
207
+ },
208
+ {
209
+ "epoch": 8.52,
210
+ "learning_rate": 5.761182437712377e-06,
211
+ "loss": 0.0011,
212
+ "step": 9500
213
+ },
214
+ {
215
+ "epoch": 8.74,
216
+ "eval_f1": 0.9957135202714771,
217
+ "eval_loss": 0.0006867917836643755,
218
+ "eval_runtime": 17.1265,
219
+ "eval_samples_per_second": 520.596,
220
+ "eval_steps_per_second": 65.104,
221
+ "step": 9746
222
+ },
223
+ {
224
+ "epoch": 8.97,
225
+ "learning_rate": 4.015369577799535e-06,
226
+ "loss": 0.002,
227
+ "step": 10000
228
+ },
229
+ {
230
+ "epoch": 9.42,
231
+ "learning_rate": 2.269556717886694e-06,
232
+ "loss": 0.0014,
233
+ "step": 10500
234
+ },
235
+ {
236
+ "epoch": 9.54,
237
+ "eval_f1": 0.9974986599964266,
238
+ "eval_loss": 0.0003464366600383073,
239
+ "eval_runtime": 17.2334,
240
+ "eval_samples_per_second": 517.369,
241
+ "eval_steps_per_second": 64.7,
242
+ "step": 10632
243
+ }
244
+ ],
245
+ "max_steps": 11150,
246
+ "num_train_epochs": 10,
247
+ "total_flos": 2805348170687904.0,
248
+ "trial_name": null,
249
+ "trial_params": {
250
+ "adam_epsilon": 7.614260573309694e-10,
251
+ "learning_rate": 3.890369377029776e-05,
252
+ "num_train_epochs": 10,
253
+ "per_device_eval_batch_size": 8,
254
+ "per_device_train_batch_size": 8,
255
+ "seed": 326,
256
+ "warmup_steps": 8,
257
+ "weight_decay": 2.2273844202742368e-07
258
+ }
259
+ }