duyvt6663 commited on
Commit
5042f1b
1 Parent(s): 3cf0384

Training in progress, step 900, checkpoint

Browse files
checkpoint-900/adapter_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c93ec6817c3a807d0df4e114c670e9b2428a5369290cdcfffa060a75ddb3fd1b
3
  size 9873829
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d818ec7785c882f3ec091517090c00e68b189be32271e7e97370e76eb2e7a5c0
3
  size 9873829
checkpoint-900/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:04d8383783fbe19d3e09463412fb03a3a9b70f027524d82824792a523d71e511
3
  size 42788
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:51345971ae33586234ce056197534c91b0075531fe4cb685ee3d82d1f64207d7
3
  size 42788
checkpoint-900/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8abe39554e2c9a5dc483956258f36b157976d41eb982b93a2108a34065d8202e
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ec6190ce24facefd0e04f00c780e140ada3c8235428d478a5b11020cf72a76ba
3
  size 14244
checkpoint-900/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:08f08011d1fc94a2dd8ae2a17ee2d6cc662d90c0383e69c3b1f8a79e789f1ad8
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4082049c84276255bb5d90bd0ce0bc02bdf34e59173f327a5829692e31d01bf8
3
  size 1064
checkpoint-900/trainer_state.json CHANGED
@@ -1,7 +1,7 @@
1
  {
2
- "best_metric": 0.6053406596183777,
3
- "best_model_checkpoint": "output/checkpoint-500",
4
- "epoch": 0.721659817580435,
5
  "eval_steps": 50,
6
  "global_step": 900,
7
  "is_hyper_param_search": false,
@@ -10,277 +10,277 @@
10
  "log_history": [
11
  {
12
  "epoch": 0.0,
13
- "learning_rate": 1.3333333333333332e-06,
14
- "loss": 0.6133,
15
  "step": 1
16
  },
17
  {
18
  "epoch": 0.04,
19
- "learning_rate": 5.9999999999999995e-05,
20
- "loss": 0.61,
21
  "step": 50
22
  },
23
  {
24
  "epoch": 0.04,
25
- "eval_accuracy": 0.6895522388059702,
26
- "eval_loss": 0.6718239784240723,
27
- "eval_runtime": 61.4305,
28
- "eval_samples_per_second": 5.453,
29
- "eval_steps_per_second": 1.367,
30
  "step": 50
31
  },
32
  {
33
  "epoch": 0.08,
34
- "learning_rate": 0.00012666666666666666,
35
- "loss": 0.6233,
36
  "step": 100
37
  },
38
  {
39
  "epoch": 0.08,
40
- "eval_accuracy": 0.6835820895522388,
41
- "eval_loss": 0.695776641368866,
42
- "eval_runtime": 61.5801,
43
- "eval_samples_per_second": 5.44,
44
- "eval_steps_per_second": 1.364,
45
  "step": 100
46
  },
47
  {
48
  "epoch": 0.12,
49
- "learning_rate": 0.00019333333333333333,
50
- "loss": 0.6187,
51
  "step": 150
52
  },
53
  {
54
  "epoch": 0.12,
55
- "eval_accuracy": 0.6895522388059702,
56
- "eval_loss": 0.6547008156776428,
57
- "eval_runtime": 61.4876,
58
- "eval_samples_per_second": 5.448,
59
- "eval_steps_per_second": 1.366,
60
  "step": 150
61
  },
62
  {
63
  "epoch": 0.16,
64
- "learning_rate": 0.00026,
65
- "loss": 0.667,
66
  "step": 200
67
  },
68
  {
69
  "epoch": 0.16,
70
- "eval_accuracy": 0.6686567164179105,
71
- "eval_loss": 0.7339211702346802,
72
- "eval_runtime": 61.0301,
73
- "eval_samples_per_second": 5.489,
74
- "eval_steps_per_second": 1.376,
75
  "step": 200
76
  },
77
  {
78
  "epoch": 0.2,
79
- "learning_rate": 0.0002941291585127201,
80
- "loss": 0.6896,
81
  "step": 250
82
  },
83
  {
84
  "epoch": 0.2,
85
- "eval_accuracy": 0.6388059701492538,
86
- "eval_loss": 0.6455020904541016,
87
- "eval_runtime": 61.0356,
88
- "eval_samples_per_second": 5.489,
89
- "eval_steps_per_second": 1.376,
90
  "step": 250
91
  },
92
  {
93
  "epoch": 0.24,
94
- "learning_rate": 0.0002794520547945205,
95
- "loss": 0.6732,
96
  "step": 300
97
  },
98
  {
99
  "epoch": 0.24,
100
- "eval_accuracy": 0.6388059701492538,
101
- "eval_loss": 0.8866483569145203,
102
- "eval_runtime": 61.5006,
103
- "eval_samples_per_second": 5.447,
104
- "eval_steps_per_second": 1.366,
105
  "step": 300
106
  },
107
  {
108
  "epoch": 0.28,
109
- "learning_rate": 0.0002647749510763209,
110
- "loss": 0.6255,
111
  "step": 350
112
  },
113
  {
114
  "epoch": 0.28,
115
- "eval_accuracy": 0.6746268656716418,
116
- "eval_loss": 0.6250954866409302,
117
- "eval_runtime": 61.4499,
118
- "eval_samples_per_second": 5.452,
119
- "eval_steps_per_second": 1.367,
120
  "step": 350
121
  },
122
  {
123
  "epoch": 0.32,
124
- "learning_rate": 0.0002500978473581213,
125
- "loss": 0.6522,
126
  "step": 400
127
  },
128
  {
129
  "epoch": 0.32,
130
- "eval_accuracy": 0.6925373134328359,
131
- "eval_loss": 0.7217190861701965,
132
- "eval_runtime": 61.1569,
133
- "eval_samples_per_second": 5.478,
134
- "eval_steps_per_second": 1.374,
135
  "step": 400
136
  },
137
  {
138
  "epoch": 0.36,
139
- "learning_rate": 0.00023542074363992171,
140
- "loss": 0.6592,
141
  "step": 450
142
  },
143
  {
144
  "epoch": 0.36,
145
- "eval_accuracy": 0.6417910447761194,
146
- "eval_loss": 0.636791467666626,
147
- "eval_runtime": 61.431,
148
- "eval_samples_per_second": 5.453,
149
- "eval_steps_per_second": 1.367,
150
  "step": 450
151
  },
152
  {
153
  "epoch": 0.4,
154
- "learning_rate": 0.00022074363992172208,
155
- "loss": 0.624,
156
  "step": 500
157
  },
158
  {
159
  "epoch": 0.4,
160
- "eval_accuracy": 0.7044776119402985,
161
- "eval_loss": 0.6053406596183777,
162
- "eval_runtime": 61.4573,
163
- "eval_samples_per_second": 5.451,
164
- "eval_steps_per_second": 1.367,
165
  "step": 500
166
  },
167
  {
168
  "epoch": 0.44,
169
- "learning_rate": 0.0002060665362035225,
170
- "loss": 0.6098,
171
  "step": 550
172
  },
173
  {
174
  "epoch": 0.44,
175
- "eval_accuracy": 0.6477611940298508,
176
- "eval_loss": 0.8857873678207397,
177
- "eval_runtime": 61.4645,
178
- "eval_samples_per_second": 5.45,
179
- "eval_steps_per_second": 1.367,
180
  "step": 550
181
  },
182
  {
183
  "epoch": 0.48,
184
- "learning_rate": 0.00019138943248532287,
185
- "loss": 0.6357,
186
  "step": 600
187
  },
188
  {
189
  "epoch": 0.48,
190
- "eval_accuracy": 0.7074626865671642,
191
- "eval_loss": 0.607388436794281,
192
- "eval_runtime": 61.4941,
193
- "eval_samples_per_second": 5.448,
194
- "eval_steps_per_second": 1.366,
195
  "step": 600
196
  },
197
  {
198
  "epoch": 0.52,
199
- "learning_rate": 0.0001767123287671233,
200
- "loss": 0.6103,
201
  "step": 650
202
  },
203
  {
204
  "epoch": 0.52,
205
- "eval_accuracy": 0.6746268656716418,
206
- "eval_loss": 0.6098934412002563,
207
- "eval_runtime": 56.4929,
208
- "eval_samples_per_second": 5.93,
209
- "eval_steps_per_second": 1.487,
210
  "step": 650
211
  },
212
  {
213
  "epoch": 0.56,
214
- "learning_rate": 0.00016203522504892366,
215
- "loss": 0.6025,
216
  "step": 700
217
  },
218
  {
219
  "epoch": 0.56,
220
- "eval_accuracy": 0.6895522388059702,
221
- "eval_loss": 0.6117912530899048,
222
- "eval_runtime": 61.1994,
223
- "eval_samples_per_second": 5.474,
224
- "eval_steps_per_second": 1.373,
225
  "step": 700
226
  },
227
  {
228
  "epoch": 0.6,
229
- "learning_rate": 0.00014735812133072405,
230
- "loss": 0.5837,
231
  "step": 750
232
  },
233
  {
234
  "epoch": 0.6,
235
- "eval_accuracy": 0.6925373134328359,
236
- "eval_loss": 0.6470033526420593,
237
- "eval_runtime": 61.3513,
238
- "eval_samples_per_second": 5.46,
239
- "eval_steps_per_second": 1.369,
240
  "step": 750
241
  },
242
  {
243
  "epoch": 0.64,
244
- "learning_rate": 0.00013268101761252445,
245
- "loss": 0.623,
246
  "step": 800
247
  },
248
  {
249
  "epoch": 0.64,
250
- "eval_accuracy": 0.6776119402985075,
251
- "eval_loss": 0.7178429365158081,
252
- "eval_runtime": 56.5122,
253
- "eval_samples_per_second": 5.928,
254
- "eval_steps_per_second": 1.486,
255
  "step": 800
256
  },
257
  {
258
  "epoch": 0.68,
259
- "learning_rate": 0.00011800391389432484,
260
- "loss": 0.6285,
261
  "step": 850
262
  },
263
  {
264
  "epoch": 0.68,
265
- "eval_accuracy": 0.6805970149253732,
266
- "eval_loss": 0.7012776732444763,
267
- "eval_runtime": 61.0309,
268
- "eval_samples_per_second": 5.489,
269
- "eval_steps_per_second": 1.376,
270
  "step": 850
271
  },
272
  {
273
  "epoch": 0.72,
274
- "learning_rate": 0.00010332681017612522,
275
- "loss": 0.609,
276
  "step": 900
277
  },
278
  {
279
  "epoch": 0.72,
280
  "eval_accuracy": 0.6656716417910448,
281
- "eval_loss": 0.6416519284248352,
282
- "eval_runtime": 56.5079,
283
- "eval_samples_per_second": 5.928,
284
  "eval_steps_per_second": 1.487,
285
  "step": 900
286
  }
@@ -288,8 +288,8 @@
288
  "logging_steps": 50,
289
  "max_steps": 1247,
290
  "num_train_epochs": 1,
291
- "save_steps": 100,
292
- "total_flos": 3.375538422344909e+17,
293
  "trial_name": null,
294
  "trial_params": null
295
  }
 
1
  {
2
+ "best_metric": 0.6430063247680664,
3
+ "best_model_checkpoint": "output/checkpoint-50",
4
+ "epoch": 0.7215151818819521,
5
  "eval_steps": 50,
6
  "global_step": 900,
7
  "is_hyper_param_search": false,
 
10
  "log_history": [
11
  {
12
  "epoch": 0.0,
13
+ "learning_rate": 4.444444444444445e-07,
14
+ "loss": 0.5542,
15
  "step": 1
16
  },
17
  {
18
  "epoch": 0.04,
19
+ "learning_rate": 2e-05,
20
+ "loss": 0.554,
21
  "step": 50
22
  },
23
  {
24
  "epoch": 0.04,
25
+ "eval_accuracy": 0.6626865671641791,
26
+ "eval_loss": 0.6430063247680664,
27
+ "eval_runtime": 61.7692,
28
+ "eval_samples_per_second": 5.423,
29
+ "eval_steps_per_second": 1.36,
30
  "step": 50
31
  },
32
  {
33
  "epoch": 0.08,
34
+ "learning_rate": 4.222222222222222e-05,
35
+ "loss": 0.5957,
36
  "step": 100
37
  },
38
  {
39
  "epoch": 0.08,
40
+ "eval_accuracy": 0.6686567164179105,
41
+ "eval_loss": 0.6705919504165649,
42
+ "eval_runtime": 60.4388,
43
+ "eval_samples_per_second": 5.543,
44
+ "eval_steps_per_second": 1.39,
45
  "step": 100
46
  },
47
  {
48
  "epoch": 0.12,
49
+ "learning_rate": 6.444444444444446e-05,
50
+ "loss": 0.5668,
51
  "step": 150
52
  },
53
  {
54
  "epoch": 0.12,
55
+ "eval_accuracy": 0.6805970149253732,
56
+ "eval_loss": 0.7224913835525513,
57
+ "eval_runtime": 56.4882,
58
+ "eval_samples_per_second": 5.93,
59
+ "eval_steps_per_second": 1.487,
60
  "step": 150
61
  },
62
  {
63
  "epoch": 0.16,
64
+ "learning_rate": 8.666666666666667e-05,
65
+ "loss": 0.5626,
66
  "step": 200
67
  },
68
  {
69
  "epoch": 0.16,
70
+ "eval_accuracy": 0.6746268656716418,
71
+ "eval_loss": 0.7115103006362915,
72
+ "eval_runtime": 56.506,
73
+ "eval_samples_per_second": 5.929,
74
+ "eval_steps_per_second": 1.487,
75
  "step": 200
76
  },
77
  {
78
  "epoch": 0.2,
79
+ "learning_rate": 9.80430528375734e-05,
80
+ "loss": 0.5604,
81
  "step": 250
82
  },
83
  {
84
  "epoch": 0.2,
85
+ "eval_accuracy": 0.6746268656716418,
86
+ "eval_loss": 0.700947105884552,
87
+ "eval_runtime": 56.4927,
88
+ "eval_samples_per_second": 5.93,
89
+ "eval_steps_per_second": 1.487,
90
  "step": 250
91
  },
92
  {
93
  "epoch": 0.24,
94
+ "learning_rate": 9.315068493150684e-05,
95
+ "loss": 0.5675,
96
  "step": 300
97
  },
98
  {
99
  "epoch": 0.24,
100
+ "eval_accuracy": 0.6597014925373135,
101
+ "eval_loss": 0.662155032157898,
102
+ "eval_runtime": 56.4903,
103
+ "eval_samples_per_second": 5.93,
104
+ "eval_steps_per_second": 1.487,
105
  "step": 300
106
  },
107
  {
108
  "epoch": 0.28,
109
+ "learning_rate": 8.825831702544032e-05,
110
+ "loss": 0.5725,
111
  "step": 350
112
  },
113
  {
114
  "epoch": 0.28,
115
+ "eval_accuracy": 0.6567164179104478,
116
+ "eval_loss": 0.6746364831924438,
117
+ "eval_runtime": 56.482,
118
+ "eval_samples_per_second": 5.931,
119
+ "eval_steps_per_second": 1.487,
120
  "step": 350
121
  },
122
  {
123
  "epoch": 0.32,
124
+ "learning_rate": 8.336594911937378e-05,
125
+ "loss": 0.5884,
126
  "step": 400
127
  },
128
  {
129
  "epoch": 0.32,
130
+ "eval_accuracy": 0.6656716417910448,
131
+ "eval_loss": 0.7047849893569946,
132
+ "eval_runtime": 56.4873,
133
+ "eval_samples_per_second": 5.931,
134
+ "eval_steps_per_second": 1.487,
135
  "step": 400
136
  },
137
  {
138
  "epoch": 0.36,
139
+ "learning_rate": 7.847358121330725e-05,
140
+ "loss": 0.5567,
141
  "step": 450
142
  },
143
  {
144
  "epoch": 0.36,
145
+ "eval_accuracy": 0.6477611940298508,
146
+ "eval_loss": 0.6832321286201477,
147
+ "eval_runtime": 56.4814,
148
+ "eval_samples_per_second": 5.931,
149
+ "eval_steps_per_second": 1.487,
150
  "step": 450
151
  },
152
  {
153
  "epoch": 0.4,
154
+ "learning_rate": 7.35812133072407e-05,
155
+ "loss": 0.5723,
156
  "step": 500
157
  },
158
  {
159
  "epoch": 0.4,
160
+ "eval_accuracy": 0.6955223880597015,
161
+ "eval_loss": 0.7437570691108704,
162
+ "eval_runtime": 56.4677,
163
+ "eval_samples_per_second": 5.933,
164
+ "eval_steps_per_second": 1.488,
165
  "step": 500
166
  },
167
  {
168
  "epoch": 0.44,
169
+ "learning_rate": 6.868884540117417e-05,
170
+ "loss": 0.5908,
171
  "step": 550
172
  },
173
  {
174
  "epoch": 0.44,
175
+ "eval_accuracy": 0.6238805970149254,
176
+ "eval_loss": 0.6612433791160583,
177
+ "eval_runtime": 56.4799,
178
+ "eval_samples_per_second": 5.931,
179
+ "eval_steps_per_second": 1.487,
180
  "step": 550
181
  },
182
  {
183
  "epoch": 0.48,
184
+ "learning_rate": 6.379647749510763e-05,
185
+ "loss": 0.5681,
186
  "step": 600
187
  },
188
  {
189
  "epoch": 0.48,
190
+ "eval_accuracy": 0.6507462686567164,
191
+ "eval_loss": 0.6833732724189758,
192
+ "eval_runtime": 56.4689,
193
+ "eval_samples_per_second": 5.932,
194
+ "eval_steps_per_second": 1.488,
195
  "step": 600
196
  },
197
  {
198
  "epoch": 0.52,
199
+ "learning_rate": 5.89041095890411e-05,
200
+ "loss": 0.5782,
201
  "step": 650
202
  },
203
  {
204
  "epoch": 0.52,
205
+ "eval_accuracy": 0.6507462686567164,
206
+ "eval_loss": 0.67350834608078,
207
+ "eval_runtime": 56.4637,
208
+ "eval_samples_per_second": 5.933,
209
+ "eval_steps_per_second": 1.488,
210
  "step": 650
211
  },
212
  {
213
  "epoch": 0.56,
214
+ "learning_rate": 5.401174168297456e-05,
215
+ "loss": 0.569,
216
  "step": 700
217
  },
218
  {
219
  "epoch": 0.56,
220
+ "eval_accuracy": 0.6686567164179105,
221
+ "eval_loss": 0.7085319757461548,
222
+ "eval_runtime": 56.4627,
223
+ "eval_samples_per_second": 5.933,
224
+ "eval_steps_per_second": 1.488,
225
  "step": 700
226
  },
227
  {
228
  "epoch": 0.6,
229
+ "learning_rate": 4.911937377690802e-05,
230
+ "loss": 0.5661,
231
  "step": 750
232
  },
233
  {
234
  "epoch": 0.6,
235
+ "eval_accuracy": 0.6835820895522388,
236
+ "eval_loss": 0.7564846277236938,
237
+ "eval_runtime": 56.4648,
238
+ "eval_samples_per_second": 5.933,
239
+ "eval_steps_per_second": 1.488,
240
  "step": 750
241
  },
242
  {
243
  "epoch": 0.64,
244
+ "learning_rate": 4.422700587084149e-05,
245
+ "loss": 0.5675,
246
  "step": 800
247
  },
248
  {
249
  "epoch": 0.64,
250
+ "eval_accuracy": 0.6417910447761194,
251
+ "eval_loss": 0.6759604215621948,
252
+ "eval_runtime": 56.4652,
253
+ "eval_samples_per_second": 5.933,
254
+ "eval_steps_per_second": 1.488,
255
  "step": 800
256
  },
257
  {
258
  "epoch": 0.68,
259
+ "learning_rate": 3.933463796477495e-05,
260
+ "loss": 0.5611,
261
  "step": 850
262
  },
263
  {
264
  "epoch": 0.68,
265
+ "eval_accuracy": 0.6477611940298508,
266
+ "eval_loss": 0.6798116564750671,
267
+ "eval_runtime": 56.4714,
268
+ "eval_samples_per_second": 5.932,
269
+ "eval_steps_per_second": 1.487,
270
  "step": 850
271
  },
272
  {
273
  "epoch": 0.72,
274
+ "learning_rate": 3.4442270058708414e-05,
275
+ "loss": 0.5664,
276
  "step": 900
277
  },
278
  {
279
  "epoch": 0.72,
280
  "eval_accuracy": 0.6656716417910448,
281
+ "eval_loss": 0.7185413837432861,
282
+ "eval_runtime": 56.4811,
283
+ "eval_samples_per_second": 5.931,
284
  "eval_steps_per_second": 1.487,
285
  "step": 900
286
  }
 
288
  "logging_steps": 50,
289
  "max_steps": 1247,
290
  "num_train_epochs": 1,
291
+ "save_steps": 50,
292
+ "total_flos": 3.3589965577015296e+17,
293
  "trial_name": null,
294
  "trial_params": null
295
  }
checkpoint-900/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e32e120c6d65576e6fe76fdf63d907877e27469586cc746d8cd6a71d07635ea9
3
  size 4472
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87bd30bd4b21b4a0c66f1fd545858ef6e2d33b5984a8d292e79011acc3affd6b
3
  size 4472