tmnam20 commited on
Commit
0eff60d
1 Parent(s): ca2659d

Training in progress, step 15000, checkpoint

Browse files
checkpoint-15000/added_tokens.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "<bos>": 40478,
3
+ "<eos>": 40479,
4
+ "<pad>": 40480
5
+ }
checkpoint-15000/config.json ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "openai-gpt",
3
+ "afn": "gelu",
4
+ "architectures": [
5
+ "OpenAIGPTForSequenceClassification"
6
+ ],
7
+ "attn_pdrop": 0.1,
8
+ "embd_pdrop": 0.1,
9
+ "id2label": {
10
+ "0": "entailment",
11
+ "1": "neutral",
12
+ "2": "contradiction"
13
+ },
14
+ "initializer_range": 0.02,
15
+ "label2id": {
16
+ "contradiction": 2,
17
+ "entailment": 0,
18
+ "neutral": 1
19
+ },
20
+ "layer_norm_epsilon": 1e-05,
21
+ "model_type": "openai-gpt",
22
+ "n_ctx": 512,
23
+ "n_embd": 768,
24
+ "n_head": 12,
25
+ "n_layer": 12,
26
+ "n_positions": 512,
27
+ "n_special": 0,
28
+ "pad_token_id": 40480,
29
+ "predict_special_tokens": true,
30
+ "problem_type": "single_label_classification",
31
+ "resid_pdrop": 0.1,
32
+ "summary_activation": null,
33
+ "summary_first_dropout": 0.1,
34
+ "summary_proj_to_labels": true,
35
+ "summary_type": "last",
36
+ "summary_use_proj": true,
37
+ "task_specific_params": {
38
+ "text-generation": {
39
+ "do_sample": true,
40
+ "max_length": 50
41
+ }
42
+ },
43
+ "torch_dtype": "float32",
44
+ "transformers_version": "4.33.0.dev0",
45
+ "vocab_size": 40481
46
+ }
checkpoint-15000/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-15000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:50cf0c1b5d758ceb5205608f8431621ad7c145001da5891ea2c1662e09ab6cfb
3
+ size 932401285
checkpoint-15000/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ee91cc89d68e288e5f20afe63584f8cb19c443853b4cb1746222b6df120ff010
3
+ size 466205145
checkpoint-15000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fea505d3bb2cc8e103041c009ac11cec0d896dbfe4c76b96d22f081a5d99d8b8
3
+ size 14575
checkpoint-15000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b4b33c6989ec640db67ac64ffd8c3b5b7340234a656f42a87bafeaadb08649a
3
+ size 627
checkpoint-15000/special_tokens_map.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<bos>",
3
+ "eos_token": "<eos>",
4
+ "pad_token": "<pad>",
5
+ "unk_token": "<unk>"
6
+ }
checkpoint-15000/tokenizer_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "clean_up_tokenization_spaces": true,
3
+ "model_max_length": 512,
4
+ "padding_side": "left",
5
+ "tokenizer_class": "OpenAIGPTTokenizer",
6
+ "unk_token": "<unk>"
7
+ }
checkpoint-15000/trainer_state.json ADDED
@@ -0,0 +1,250 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 1.1149991750717163,
3
+ "best_model_checkpoint": "/kaggle/output/checkpoint-9000",
4
+ "epoch": 240.0,
5
+ "eval_steps": 1000,
6
+ "global_step": 15000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.02,
13
+ "learning_rate": 2.7777777777777777e-11,
14
+ "loss": 1.2395,
15
+ "step": 1
16
+ },
17
+ {
18
+ "epoch": 16.0,
19
+ "learning_rate": 2.7638888888888893e-08,
20
+ "loss": 1.1422,
21
+ "step": 1000
22
+ },
23
+ {
24
+ "epoch": 16.0,
25
+ "eval_accuracy": 0.33073852295409184,
26
+ "eval_loss": 1.1557353734970093,
27
+ "eval_runtime": 54.1805,
28
+ "eval_samples_per_second": 92.469,
29
+ "eval_steps_per_second": 11.572,
30
+ "step": 1000
31
+ },
32
+ {
33
+ "epoch": 32.0,
34
+ "learning_rate": 5.541666666666667e-08,
35
+ "loss": 1.1401,
36
+ "step": 2000
37
+ },
38
+ {
39
+ "epoch": 32.0,
40
+ "eval_accuracy": 0.3311377245508982,
41
+ "eval_loss": 1.1507551670074463,
42
+ "eval_runtime": 53.98,
43
+ "eval_samples_per_second": 92.812,
44
+ "eval_steps_per_second": 11.615,
45
+ "step": 2000
46
+ },
47
+ {
48
+ "epoch": 48.0,
49
+ "learning_rate": 8.316666666666666e-08,
50
+ "loss": 1.1287,
51
+ "step": 3000
52
+ },
53
+ {
54
+ "epoch": 48.0,
55
+ "eval_accuracy": 0.3321357285429142,
56
+ "eval_loss": 1.1441665887832642,
57
+ "eval_runtime": 53.9873,
58
+ "eval_samples_per_second": 92.8,
59
+ "eval_steps_per_second": 11.614,
60
+ "step": 3000
61
+ },
62
+ {
63
+ "epoch": 64.0,
64
+ "learning_rate": 1.1094444444444445e-07,
65
+ "loss": 1.117,
66
+ "step": 4000
67
+ },
68
+ {
69
+ "epoch": 64.0,
70
+ "eval_accuracy": 0.3287425149700599,
71
+ "eval_loss": 1.134506344795227,
72
+ "eval_runtime": 54.0067,
73
+ "eval_samples_per_second": 92.766,
74
+ "eval_steps_per_second": 11.61,
75
+ "step": 4000
76
+ },
77
+ {
78
+ "epoch": 80.0,
79
+ "learning_rate": 1.3872222222222222e-07,
80
+ "loss": 1.1111,
81
+ "step": 5000
82
+ },
83
+ {
84
+ "epoch": 80.0,
85
+ "eval_accuracy": 0.32954091816367265,
86
+ "eval_loss": 1.1283142566680908,
87
+ "eval_runtime": 54.0921,
88
+ "eval_samples_per_second": 92.62,
89
+ "eval_steps_per_second": 11.591,
90
+ "step": 5000
91
+ },
92
+ {
93
+ "epoch": 96.0,
94
+ "learning_rate": 1.6650000000000002e-07,
95
+ "loss": 1.1053,
96
+ "step": 6000
97
+ },
98
+ {
99
+ "epoch": 96.0,
100
+ "eval_accuracy": 0.32894211576846305,
101
+ "eval_loss": 1.1218452453613281,
102
+ "eval_runtime": 54.0607,
103
+ "eval_samples_per_second": 92.674,
104
+ "eval_steps_per_second": 11.598,
105
+ "step": 6000
106
+ },
107
+ {
108
+ "epoch": 112.0,
109
+ "learning_rate": 1.9422222222222223e-07,
110
+ "loss": 1.0952,
111
+ "step": 7000
112
+ },
113
+ {
114
+ "epoch": 112.0,
115
+ "eval_accuracy": 0.33073852295409184,
116
+ "eval_loss": 1.118302583694458,
117
+ "eval_runtime": 53.9252,
118
+ "eval_samples_per_second": 92.906,
119
+ "eval_steps_per_second": 11.627,
120
+ "step": 7000
121
+ },
122
+ {
123
+ "epoch": 128.0,
124
+ "learning_rate": 2.2197222222222225e-07,
125
+ "loss": 1.0909,
126
+ "step": 8000
127
+ },
128
+ {
129
+ "epoch": 128.0,
130
+ "eval_accuracy": 0.3281437125748503,
131
+ "eval_loss": 1.1167163848876953,
132
+ "eval_runtime": 54.095,
133
+ "eval_samples_per_second": 92.615,
134
+ "eval_steps_per_second": 11.591,
135
+ "step": 8000
136
+ },
137
+ {
138
+ "epoch": 144.0,
139
+ "learning_rate": 2.4975e-07,
140
+ "loss": 1.0838,
141
+ "step": 9000
142
+ },
143
+ {
144
+ "epoch": 144.0,
145
+ "eval_accuracy": 0.3317365269461078,
146
+ "eval_loss": 1.1149991750717163,
147
+ "eval_runtime": 54.0536,
148
+ "eval_samples_per_second": 92.686,
149
+ "eval_steps_per_second": 11.6,
150
+ "step": 9000
151
+ },
152
+ {
153
+ "epoch": 160.0,
154
+ "learning_rate": 2.775277777777778e-07,
155
+ "loss": 1.0746,
156
+ "step": 10000
157
+ },
158
+ {
159
+ "epoch": 160.0,
160
+ "eval_accuracy": 0.3347305389221557,
161
+ "eval_loss": 1.1178398132324219,
162
+ "eval_runtime": 54.1273,
163
+ "eval_samples_per_second": 92.56,
164
+ "eval_steps_per_second": 11.584,
165
+ "step": 10000
166
+ },
167
+ {
168
+ "epoch": 176.0,
169
+ "learning_rate": 3.0530555555555556e-07,
170
+ "loss": 1.0722,
171
+ "step": 11000
172
+ },
173
+ {
174
+ "epoch": 176.0,
175
+ "eval_accuracy": 0.3373253493013972,
176
+ "eval_loss": 1.1214321851730347,
177
+ "eval_runtime": 54.046,
178
+ "eval_samples_per_second": 92.699,
179
+ "eval_steps_per_second": 11.601,
180
+ "step": 11000
181
+ },
182
+ {
183
+ "epoch": 192.0,
184
+ "learning_rate": 3.3308333333333333e-07,
185
+ "loss": 1.0617,
186
+ "step": 12000
187
+ },
188
+ {
189
+ "epoch": 192.0,
190
+ "eval_accuracy": 0.34271457085828344,
191
+ "eval_loss": 1.124221682548523,
192
+ "eval_runtime": 54.0489,
193
+ "eval_samples_per_second": 92.694,
194
+ "eval_steps_per_second": 11.601,
195
+ "step": 12000
196
+ },
197
+ {
198
+ "epoch": 208.0,
199
+ "learning_rate": 3.608611111111111e-07,
200
+ "loss": 1.0509,
201
+ "step": 13000
202
+ },
203
+ {
204
+ "epoch": 208.0,
205
+ "eval_accuracy": 0.3475049900199601,
206
+ "eval_loss": 1.1377869844436646,
207
+ "eval_runtime": 54.0137,
208
+ "eval_samples_per_second": 92.754,
209
+ "eval_steps_per_second": 11.608,
210
+ "step": 13000
211
+ },
212
+ {
213
+ "epoch": 224.0,
214
+ "learning_rate": 3.8863888888888893e-07,
215
+ "loss": 1.0381,
216
+ "step": 14000
217
+ },
218
+ {
219
+ "epoch": 224.0,
220
+ "eval_accuracy": 0.34890219560878244,
221
+ "eval_loss": 1.155613899230957,
222
+ "eval_runtime": 53.9794,
223
+ "eval_samples_per_second": 92.813,
224
+ "eval_steps_per_second": 11.616,
225
+ "step": 14000
226
+ },
227
+ {
228
+ "epoch": 240.0,
229
+ "learning_rate": 4.163888888888889e-07,
230
+ "loss": 1.0255,
231
+ "step": 15000
232
+ },
233
+ {
234
+ "epoch": 240.0,
235
+ "eval_accuracy": 0.3562874251497006,
236
+ "eval_loss": 1.1687414646148682,
237
+ "eval_runtime": 53.8948,
238
+ "eval_samples_per_second": 92.959,
239
+ "eval_steps_per_second": 11.634,
240
+ "step": 15000
241
+ }
242
+ ],
243
+ "logging_steps": 1000,
244
+ "max_steps": 10000000,
245
+ "num_train_epochs": 161291,
246
+ "save_steps": 1000,
247
+ "total_flos": 3.135532695552e+16,
248
+ "trial_name": null,
249
+ "trial_params": null
250
+ }
checkpoint-15000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cb2cd1b7e648a386ccb79e47a451842e11001981c4363e993b678e0818db5cab
3
+ size 4091
checkpoint-15000/vocab.json ADDED
The diff for this file is too large to render. See raw diff