Hugging Face's logo

tyzhu
/
lmind_nq_train600_eval300_v1_docidx_gpt2-xl

Text Generation
Transformers
PyTorch
gpt2
Generated from Trainer
Eval Results
text-generation-inference
Model card Files Community
1
tyzhu commited on
Commit
e403ea5
·
verified ·
1 Parent(s): 680264a

Training in progress, epoch 3, checkpoint

Browse files
Files changed (14) hide show
  1. checkpoint-168/added_tokens.json +3 -0
  2. checkpoint-168/config.json +42 -0
  3. checkpoint-168/generation_config.json +8 -0
  4. checkpoint-168/merges.txt +0 -0
  5. checkpoint-168/optimizer.pt +3 -0
  6. checkpoint-168/pytorch_model.bin +3 -0
  7. checkpoint-168/rng_state.pth +3 -0
  8. checkpoint-168/scheduler.pt +3 -0
  9. checkpoint-168/special_tokens_map.json +6 -0
  10. checkpoint-168/tokenizer.json +0 -0
  11. checkpoint-168/tokenizer_config.json +20 -0
  12. checkpoint-168/trainer_state.json +283 -0
  13. checkpoint-168/training_args.bin +3 -0
  14. checkpoint-168/vocab.json +0 -0
checkpoint-168/added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "<|endoftext|>": 50256
3
+ }
checkpoint-168/config.json ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "gpt2-xl",
3
+ "activation_function": "gelu_new",
4
+ "architectures": [
5
+ "GPT2LMHeadModel"
6
+ ],
7
+ "attn_pdrop": 0.1,
8
+ "bos_token_id": 50256,
9
+ "do_sample": true,
10
+ "embd_pdrop": 0.1,
11
+ "eos_token_id": 50256,
12
+ "initializer_range": 0.02,
13
+ "layer_norm_epsilon": 1e-05,
14
+ "max_length": 50,
15
+ "model_type": "gpt2",
16
+ "n_ctx": 1024,
17
+ "n_embd": 1600,
18
+ "n_head": 25,
19
+ "n_inner": null,
20
+ "n_layer": 48,
21
+ "n_positions": 1024,
22
+ "output_past": true,
23
+ "reorder_and_upcast_attn": false,
24
+ "resid_pdrop": 0.1,
25
+ "scale_attn_by_inverse_layer_idx": false,
26
+ "scale_attn_weights": true,
27
+ "summary_activation": null,
28
+ "summary_first_dropout": 0.1,
29
+ "summary_proj_to_labels": true,
30
+ "summary_type": "cls_index",
31
+ "summary_use_proj": true,
32
+ "task_specific_params": {
33
+ "text-generation": {
34
+ "do_sample": true,
35
+ "max_length": 50
36
+ }
37
+ },
38
+ "torch_dtype": "float32",
39
+ "transformers_version": "4.34.0",
40
+ "use_cache": true,
41
+ "vocab_size": 50257
42
+ }
checkpoint-168/generation_config.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 50256,
4
+ "do_sample": true,
5
+ "eos_token_id": 50256,
6
+ "max_length": 50,
7
+ "transformers_version": "4.34.0"
8
+ }
checkpoint-168/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-168/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42847bcd4085a3b4fd9c8fd4ba5be01e1b668a1b0f4c0ee700800168c9b7ccb2
3
+ size 12461385454
checkpoint-168/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:adb58da854628dd03d59c22938b09ed0a4215bfb916e881dfd0d812600c50bcb
3
+ size 6230637102
checkpoint-168/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c3199ebbae246ff1c79c3634c9d9052490675bf051c616cb408b99adde733e9d
3
+ size 14244
checkpoint-168/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:68228e9738753ecb60c6552797935f09d92a791c98f66745bc4fbe8eef412d17
3
+ size 1064
checkpoint-168/special_tokens_map.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<|endoftext|>",
3
+ "eos_token": "<|endoftext|>",
4
+ "pad_token": "<|endoftext|>",
5
+ "unk_token": "<|endoftext|>"
6
+ }
checkpoint-168/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-168/tokenizer_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "50256": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ }
12
+ },
13
+ "additional_special_tokens": [],
14
+ "bos_token": "<|endoftext|>",
15
+ "clean_up_tokenization_spaces": true,
16
+ "eos_token": "<|endoftext|>",
17
+ "model_max_length": 1024,
18
+ "tokenizer_class": "GPT2Tokenizer",
19
+ "unk_token": "<|endoftext|>"
20
+ }
checkpoint-168/trainer_state.json ADDED
@@ -0,0 +1,283 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0,
5
+ "eval_steps": 28,
6
+ "global_step": 168,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.11,
13
+ "learning_rate": 3e-05,
14
+ "loss": 3.5167,
15
+ "step": 6
16
+ },
17
+ {
18
+ "epoch": 0.21,
19
+ "learning_rate": 3e-05,
20
+ "loss": 2.7034,
21
+ "step": 12
22
+ },
23
+ {
24
+ "epoch": 0.32,
25
+ "learning_rate": 3e-05,
26
+ "loss": 2.5409,
27
+ "step": 18
28
+ },
29
+ {
30
+ "epoch": 0.43,
31
+ "learning_rate": 3e-05,
32
+ "loss": 2.4931,
33
+ "step": 24
34
+ },
35
+ {
36
+ "epoch": 0.5,
37
+ "eval_accuracy": 0.5748751060189339,
38
+ "eval_loss": 2.390355348587036,
39
+ "eval_runtime": 25.7729,
40
+ "eval_samples_per_second": 34.261,
41
+ "eval_steps_per_second": 2.173,
42
+ "step": 28
43
+ },
44
+ {
45
+ "epoch": 0.5,
46
+ "eval_bleu": 2.6290615444017718,
47
+ "eval_exact_match": 0.0,
48
+ "eval_prefix_exact_match": 0.0,
49
+ "step": 28
50
+ },
51
+ {
52
+ "epoch": 0.54,
53
+ "learning_rate": 3e-05,
54
+ "loss": 2.4944,
55
+ "step": 30
56
+ },
57
+ {
58
+ "epoch": 0.64,
59
+ "learning_rate": 3e-05,
60
+ "loss": 2.4832,
61
+ "step": 36
62
+ },
63
+ {
64
+ "epoch": 0.75,
65
+ "learning_rate": 3e-05,
66
+ "loss": 2.4336,
67
+ "step": 42
68
+ },
69
+ {
70
+ "epoch": 0.86,
71
+ "learning_rate": 3e-05,
72
+ "loss": 2.5101,
73
+ "step": 48
74
+ },
75
+ {
76
+ "epoch": 0.96,
77
+ "learning_rate": 3e-05,
78
+ "loss": 2.4699,
79
+ "step": 54
80
+ },
81
+ {
82
+ "epoch": 1.0,
83
+ "eval_accuracy": 0.6049526162776446,
84
+ "eval_loss": 2.101423740386963,
85
+ "eval_runtime": 23.9781,
86
+ "eval_samples_per_second": 36.825,
87
+ "eval_steps_per_second": 2.335,
88
+ "step": 56
89
+ },
90
+ {
91
+ "epoch": 1.0,
92
+ "eval_bleu": 2.731416805620766,
93
+ "eval_exact_match": 0.0,
94
+ "eval_prefix_exact_match": 0.0,
95
+ "step": 56
96
+ },
97
+ {
98
+ "epoch": 1.07,
99
+ "learning_rate": 3e-05,
100
+ "loss": 2.2172,
101
+ "step": 60
102
+ },
103
+ {
104
+ "epoch": 1.18,
105
+ "learning_rate": 3e-05,
106
+ "loss": 1.9462,
107
+ "step": 66
108
+ },
109
+ {
110
+ "epoch": 1.29,
111
+ "learning_rate": 3e-05,
112
+ "loss": 1.8014,
113
+ "step": 72
114
+ },
115
+ {
116
+ "epoch": 1.39,
117
+ "learning_rate": 3e-05,
118
+ "loss": 1.8689,
119
+ "step": 78
120
+ },
121
+ {
122
+ "epoch": 1.5,
123
+ "learning_rate": 3e-05,
124
+ "loss": 1.9006,
125
+ "step": 84
126
+ },
127
+ {
128
+ "epoch": 1.5,
129
+ "eval_accuracy": 0.6392953969397911,
130
+ "eval_loss": 1.8079485893249512,
131
+ "eval_runtime": 24.4889,
132
+ "eval_samples_per_second": 36.057,
133
+ "eval_steps_per_second": 2.287,
134
+ "step": 84
135
+ },
136
+ {
137
+ "epoch": 1.5,
138
+ "eval_bleu": 3.957789284907739,
139
+ "eval_exact_match": 0.0,
140
+ "eval_prefix_exact_match": 0.0,
141
+ "step": 84
142
+ },
143
+ {
144
+ "epoch": 1.61,
145
+ "learning_rate": 3e-05,
146
+ "loss": 1.8975,
147
+ "step": 90
148
+ },
149
+ {
150
+ "epoch": 1.71,
151
+ "learning_rate": 3e-05,
152
+ "loss": 1.8936,
153
+ "step": 96
154
+ },
155
+ {
156
+ "epoch": 1.82,
157
+ "learning_rate": 3e-05,
158
+ "loss": 1.8677,
159
+ "step": 102
160
+ },
161
+ {
162
+ "epoch": 1.93,
163
+ "learning_rate": 3e-05,
164
+ "loss": 1.9317,
165
+ "step": 108
166
+ },
167
+ {
168
+ "epoch": 2.0,
169
+ "eval_accuracy": 0.6722017129718149,
170
+ "eval_loss": 1.5510306358337402,
171
+ "eval_runtime": 24.2083,
172
+ "eval_samples_per_second": 36.475,
173
+ "eval_steps_per_second": 2.313,
174
+ "step": 112
175
+ },
176
+ {
177
+ "epoch": 2.0,
178
+ "eval_bleu": 4.403977979732999,
179
+ "eval_exact_match": 0.0,
180
+ "eval_prefix_exact_match": 0.0,
181
+ "step": 112
182
+ },
183
+ {
184
+ "epoch": 2.04,
185
+ "learning_rate": 3e-05,
186
+ "loss": 1.7801,
187
+ "step": 114
188
+ },
189
+ {
190
+ "epoch": 2.14,
191
+ "learning_rate": 3e-05,
192
+ "loss": 1.3602,
193
+ "step": 120
194
+ },
195
+ {
196
+ "epoch": 2.25,
197
+ "learning_rate": 3e-05,
198
+ "loss": 1.3667,
199
+ "step": 126
200
+ },
201
+ {
202
+ "epoch": 2.36,
203
+ "learning_rate": 3e-05,
204
+ "loss": 1.3497,
205
+ "step": 132
206
+ },
207
+ {
208
+ "epoch": 2.46,
209
+ "learning_rate": 3e-05,
210
+ "loss": 1.3984,
211
+ "step": 138
212
+ },
213
+ {
214
+ "epoch": 2.5,
215
+ "eval_accuracy": 0.707495599907831,
216
+ "eval_loss": 1.284958839416504,
217
+ "eval_runtime": 24.875,
218
+ "eval_samples_per_second": 35.498,
219
+ "eval_steps_per_second": 2.251,
220
+ "step": 140
221
+ },
222
+ {
223
+ "epoch": 2.5,
224
+ "eval_bleu": 7.1028643857654,
225
+ "eval_exact_match": 0.0,
226
+ "eval_prefix_exact_match": 0.0,
227
+ "step": 140
228
+ },
229
+ {
230
+ "epoch": 2.57,
231
+ "learning_rate": 3e-05,
232
+ "loss": 1.3143,
233
+ "step": 144
234
+ },
235
+ {
236
+ "epoch": 2.68,
237
+ "learning_rate": 3e-05,
238
+ "loss": 1.3887,
239
+ "step": 150
240
+ },
241
+ {
242
+ "epoch": 2.79,
243
+ "learning_rate": 3e-05,
244
+ "loss": 1.3563,
245
+ "step": 156
246
+ },
247
+ {
248
+ "epoch": 2.89,
249
+ "learning_rate": 3e-05,
250
+ "loss": 1.3749,
251
+ "step": 162
252
+ },
253
+ {
254
+ "epoch": 3.0,
255
+ "learning_rate": 3e-05,
256
+ "loss": 1.3662,
257
+ "step": 168
258
+ },
259
+ {
260
+ "epoch": 3.0,
261
+ "eval_accuracy": 0.737386810999495,
262
+ "eval_loss": 1.0899983644485474,
263
+ "eval_runtime": 24.1441,
264
+ "eval_samples_per_second": 36.572,
265
+ "eval_steps_per_second": 2.319,
266
+ "step": 168
267
+ },
268
+ {
269
+ "epoch": 3.0,
270
+ "eval_bleu": 8.181972004065782,
271
+ "eval_exact_match": 0.0,
272
+ "eval_prefix_exact_match": 0.0,
273
+ "step": 168
274
+ }
275
+ ],
276
+ "logging_steps": 6,
277
+ "max_steps": 560,
278
+ "num_train_epochs": 10,
279
+ "save_steps": 500,
280
+ "total_flos": 4091531640422400.0,
281
+ "trial_name": null,
282
+ "trial_params": null
283
+ }
checkpoint-168/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8b959b77797bfa32386725e4d78839c39c28940487e69e41fb564f8bb31a78b3
3
+ size 4664
checkpoint-168/vocab.json ADDED
The diff for this file is too large to render. See raw diff