Hugging Face's logo

tyzhu
/
lmind_hotpot_train1000_eval500_v1_recite_qa_gpt2-xl

Text Generation
Transformers
PyTorch
gpt2
Generated from Trainer
Eval Results
text-generation-inference
Model card Files Community
1
tyzhu commited on
Commit
34c1ff8
·
verified ·
1 Parent(s): 8cd0ccd

Training in progress, epoch 10, checkpoint

Browse files
Files changed (14) hide show
  1. checkpoint-2480/added_tokens.json +3 -0
  2. checkpoint-2480/config.json +42 -0
  3. checkpoint-2480/generation_config.json +8 -0
  4. checkpoint-2480/merges.txt +0 -0
  5. checkpoint-2480/optimizer.pt +3 -0
  6. checkpoint-2480/pytorch_model.bin +3 -0
  7. checkpoint-2480/rng_state.pth +3 -0
  8. checkpoint-2480/scheduler.pt +3 -0
  9. checkpoint-2480/special_tokens_map.json +6 -0
  10. checkpoint-2480/tokenizer.json +0 -0
  11. checkpoint-2480/tokenizer_config.json +20 -0
  12. checkpoint-2480/trainer_state.json +503 -0
  13. checkpoint-2480/training_args.bin +3 -0
  14. checkpoint-2480/vocab.json +0 -0
checkpoint-2480/added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "<|endoftext|>": 50256
3
+ }
checkpoint-2480/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-2480/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-2480/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-2480/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5e0dba486fdf17ecd7214ac58527c19f85edeb864672816c98a3d70b38bc0833
3
+ size 12461385454
checkpoint-2480/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a251bac6d99f4a78940f972b43fc18c1b4a4335d6d61c2ad2e4fc1ca9aedc40e
3
+ size 6230637102
checkpoint-2480/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dcc255adae61e8335645ce8854ba3789af352e6ecd506ab1080929b73cc50db4
3
+ size 14244
checkpoint-2480/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:706944265250781497fbb3ba8ae98d01c0a7039a1520931992e676d986ece7bb
3
+ size 1064
checkpoint-2480/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-2480/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-2480/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-2480/trainer_state.json ADDED
@@ -0,0 +1,503 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 10.0,
5
+ "eval_steps": 248,
6
+ "global_step": 2480,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.2,
13
+ "learning_rate": 3e-05,
14
+ "loss": 2.8428,
15
+ "step": 50
16
+ },
17
+ {
18
+ "epoch": 0.4,
19
+ "learning_rate": 3e-05,
20
+ "loss": 2.1589,
21
+ "step": 100
22
+ },
23
+ {
24
+ "epoch": 0.6,
25
+ "learning_rate": 3e-05,
26
+ "loss": 2.0527,
27
+ "step": 150
28
+ },
29
+ {
30
+ "epoch": 0.81,
31
+ "learning_rate": 3e-05,
32
+ "loss": 1.9823,
33
+ "step": 200
34
+ },
35
+ {
36
+ "epoch": 1.0,
37
+ "eval_accuracy": 0.6398457059679767,
38
+ "eval_loss": 1.6149228811264038,
39
+ "eval_runtime": 17.4504,
40
+ "eval_samples_per_second": 28.653,
41
+ "eval_steps_per_second": 1.834,
42
+ "step": 248
43
+ },
44
+ {
45
+ "epoch": 1.0,
46
+ "eval_exact_match": 9.0,
47
+ "eval_f1": 13.45666666666666,
48
+ "eval_qa_bleu": 7.063432441196736,
49
+ "eval_qa_exact_match": 0.09,
50
+ "eval_recite_bleu": 15.344658855358656,
51
+ "eval_recite_exact_match": 0.0,
52
+ "step": 248
53
+ },
54
+ {
55
+ "epoch": 1.01,
56
+ "learning_rate": 3e-05,
57
+ "loss": 1.9153,
58
+ "step": 250
59
+ },
60
+ {
61
+ "epoch": 1.21,
62
+ "learning_rate": 3e-05,
63
+ "loss": 1.49,
64
+ "step": 300
65
+ },
66
+ {
67
+ "epoch": 1.41,
68
+ "learning_rate": 3e-05,
69
+ "loss": 1.4406,
70
+ "step": 350
71
+ },
72
+ {
73
+ "epoch": 1.61,
74
+ "learning_rate": 3e-05,
75
+ "loss": 1.4161,
76
+ "step": 400
77
+ },
78
+ {
79
+ "epoch": 1.81,
80
+ "learning_rate": 3e-05,
81
+ "loss": 1.3868,
82
+ "step": 450
83
+ },
84
+ {
85
+ "epoch": 2.0,
86
+ "eval_accuracy": 0.6700320232896652,
87
+ "eval_loss": 1.1928881406784058,
88
+ "eval_runtime": 15.4486,
89
+ "eval_samples_per_second": 32.365,
90
+ "eval_steps_per_second": 2.071,
91
+ "step": 496
92
+ },
93
+ {
94
+ "epoch": 2.0,
95
+ "eval_exact_match": 9.2,
96
+ "eval_f1": 14.735098500533276,
97
+ "eval_qa_bleu": 7.255412763224127,
98
+ "eval_qa_exact_match": 0.088,
99
+ "eval_recite_bleu": 17.979965007898493,
100
+ "eval_recite_exact_match": 0.0,
101
+ "step": 496
102
+ },
103
+ {
104
+ "epoch": 2.02,
105
+ "learning_rate": 3e-05,
106
+ "loss": 1.309,
107
+ "step": 500
108
+ },
109
+ {
110
+ "epoch": 2.22,
111
+ "learning_rate": 3e-05,
112
+ "loss": 0.9401,
113
+ "step": 550
114
+ },
115
+ {
116
+ "epoch": 2.42,
117
+ "learning_rate": 3e-05,
118
+ "loss": 0.9429,
119
+ "step": 600
120
+ },
121
+ {
122
+ "epoch": 2.62,
123
+ "learning_rate": 3e-05,
124
+ "loss": 0.9115,
125
+ "step": 650
126
+ },
127
+ {
128
+ "epoch": 2.82,
129
+ "learning_rate": 3e-05,
130
+ "loss": 0.9102,
131
+ "step": 700
132
+ },
133
+ {
134
+ "epoch": 3.0,
135
+ "eval_accuracy": 0.6992489082969432,
136
+ "eval_loss": 0.8513368368148804,
137
+ "eval_runtime": 15.7421,
138
+ "eval_samples_per_second": 31.762,
139
+ "eval_steps_per_second": 2.033,
140
+ "step": 744
141
+ },
142
+ {
143
+ "epoch": 3.0,
144
+ "eval_exact_match": 10.2,
145
+ "eval_f1": 15.888450046685344,
146
+ "eval_qa_bleu": 7.2379640900667646,
147
+ "eval_qa_exact_match": 0.098,
148
+ "eval_recite_bleu": 20.876580135938546,
149
+ "eval_recite_exact_match": 0.0,
150
+ "step": 744
151
+ },
152
+ {
153
+ "epoch": 3.02,
154
+ "learning_rate": 3e-05,
155
+ "loss": 0.8302,
156
+ "step": 750
157
+ },
158
+ {
159
+ "epoch": 3.23,
160
+ "learning_rate": 3e-05,
161
+ "loss": 0.5813,
162
+ "step": 800
163
+ },
164
+ {
165
+ "epoch": 3.43,
166
+ "learning_rate": 3e-05,
167
+ "loss": 0.5844,
168
+ "step": 850
169
+ },
170
+ {
171
+ "epoch": 3.63,
172
+ "learning_rate": 3e-05,
173
+ "loss": 0.5776,
174
+ "step": 900
175
+ },
176
+ {
177
+ "epoch": 3.83,
178
+ "learning_rate": 3e-05,
179
+ "loss": 0.5869,
180
+ "step": 950
181
+ },
182
+ {
183
+ "epoch": 4.0,
184
+ "eval_accuracy": 0.721787481804949,
185
+ "eval_loss": 0.6180558204650879,
186
+ "eval_runtime": 15.5183,
187
+ "eval_samples_per_second": 32.22,
188
+ "eval_steps_per_second": 2.062,
189
+ "step": 992
190
+ },
191
+ {
192
+ "epoch": 4.0,
193
+ "eval_exact_match": 13.6,
194
+ "eval_f1": 19.457685255920552,
195
+ "eval_qa_bleu": 7.372812651620524,
196
+ "eval_qa_exact_match": 0.126,
197
+ "eval_recite_bleu": 24.119469757278573,
198
+ "eval_recite_exact_match": 0.0,
199
+ "step": 992
200
+ },
201
+ {
202
+ "epoch": 4.03,
203
+ "learning_rate": 3e-05,
204
+ "loss": 0.5394,
205
+ "step": 1000
206
+ },
207
+ {
208
+ "epoch": 4.23,
209
+ "learning_rate": 3e-05,
210
+ "loss": 0.3787,
211
+ "step": 1050
212
+ },
213
+ {
214
+ "epoch": 4.44,
215
+ "learning_rate": 3e-05,
216
+ "loss": 0.3865,
217
+ "step": 1100
218
+ },
219
+ {
220
+ "epoch": 4.64,
221
+ "learning_rate": 3e-05,
222
+ "loss": 0.3734,
223
+ "step": 1150
224
+ },
225
+ {
226
+ "epoch": 4.84,
227
+ "learning_rate": 3e-05,
228
+ "loss": 0.364,
229
+ "step": 1200
230
+ },
231
+ {
232
+ "epoch": 5.0,
233
+ "eval_accuracy": 0.7352052401746725,
234
+ "eval_loss": 0.49626532196998596,
235
+ "eval_runtime": 15.762,
236
+ "eval_samples_per_second": 31.722,
237
+ "eval_steps_per_second": 2.03,
238
+ "step": 1240
239
+ },
240
+ {
241
+ "epoch": 5.0,
242
+ "eval_exact_match": 11.6,
243
+ "eval_f1": 17.050505050505052,
244
+ "eval_qa_bleu": 9.3668737780184,
245
+ "eval_qa_exact_match": 0.114,
246
+ "eval_recite_bleu": 28.6471128712922,
247
+ "eval_recite_exact_match": 0.004,
248
+ "step": 1240
249
+ },
250
+ {
251
+ "epoch": 5.04,
252
+ "learning_rate": 3e-05,
253
+ "loss": 0.345,
254
+ "step": 1250
255
+ },
256
+ {
257
+ "epoch": 5.24,
258
+ "learning_rate": 3e-05,
259
+ "loss": 0.2508,
260
+ "step": 1300
261
+ },
262
+ {
263
+ "epoch": 5.44,
264
+ "learning_rate": 3e-05,
265
+ "loss": 0.26,
266
+ "step": 1350
267
+ },
268
+ {
269
+ "epoch": 5.65,
270
+ "learning_rate": 3e-05,
271
+ "loss": 0.2618,
272
+ "step": 1400
273
+ },
274
+ {
275
+ "epoch": 5.85,
276
+ "learning_rate": 3e-05,
277
+ "loss": 0.2706,
278
+ "step": 1450
279
+ },
280
+ {
281
+ "epoch": 6.0,
282
+ "eval_accuracy": 0.7419825327510917,
283
+ "eval_loss": 0.441346138715744,
284
+ "eval_runtime": 15.4217,
285
+ "eval_samples_per_second": 32.422,
286
+ "eval_steps_per_second": 2.075,
287
+ "step": 1488
288
+ },
289
+ {
290
+ "epoch": 6.0,
291
+ "eval_exact_match": 17.6,
292
+ "eval_f1": 23.964126984126988,
293
+ "eval_qa_bleu": 13.1735935431327,
294
+ "eval_qa_exact_match": 0.164,
295
+ "eval_recite_bleu": 35.20769002447571,
296
+ "eval_recite_exact_match": 0.008,
297
+ "step": 1488
298
+ },
299
+ {
300
+ "epoch": 6.05,
301
+ "learning_rate": 3e-05,
302
+ "loss": 0.2459,
303
+ "step": 1500
304
+ },
305
+ {
306
+ "epoch": 6.25,
307
+ "learning_rate": 3e-05,
308
+ "loss": 0.1802,
309
+ "step": 1550
310
+ },
311
+ {
312
+ "epoch": 6.45,
313
+ "learning_rate": 3e-05,
314
+ "loss": 0.1853,
315
+ "step": 1600
316
+ },
317
+ {
318
+ "epoch": 6.65,
319
+ "learning_rate": 3e-05,
320
+ "loss": 0.1905,
321
+ "step": 1650
322
+ },
323
+ {
324
+ "epoch": 6.85,
325
+ "learning_rate": 3e-05,
326
+ "loss": 0.1906,
327
+ "step": 1700
328
+ },
329
+ {
330
+ "epoch": 7.0,
331
+ "eval_accuracy": 0.7457903930131005,
332
+ "eval_loss": 0.4084840714931488,
333
+ "eval_runtime": 15.7338,
334
+ "eval_samples_per_second": 31.779,
335
+ "eval_steps_per_second": 2.034,
336
+ "step": 1736
337
+ },
338
+ {
339
+ "epoch": 7.0,
340
+ "eval_exact_match": 20.2,
341
+ "eval_f1": 25.946536796536805,
342
+ "eval_qa_bleu": 13.109935567750034,
343
+ "eval_qa_exact_match": 0.198,
344
+ "eval_recite_bleu": 41.370445416956926,
345
+ "eval_recite_exact_match": 0.018,
346
+ "step": 1736
347
+ },
348
+ {
349
+ "epoch": 7.06,
350
+ "learning_rate": 3e-05,
351
+ "loss": 0.1735,
352
+ "step": 1750
353
+ },
354
+ {
355
+ "epoch": 7.26,
356
+ "learning_rate": 3e-05,
357
+ "loss": 0.1402,
358
+ "step": 1800
359
+ },
360
+ {
361
+ "epoch": 7.46,
362
+ "learning_rate": 3e-05,
363
+ "loss": 0.1423,
364
+ "step": 1850
365
+ },
366
+ {
367
+ "epoch": 7.66,
368
+ "learning_rate": 3e-05,
369
+ "loss": 0.1468,
370
+ "step": 1900
371
+ },
372
+ {
373
+ "epoch": 7.86,
374
+ "learning_rate": 3e-05,
375
+ "loss": 0.1474,
376
+ "step": 1950
377
+ },
378
+ {
379
+ "epoch": 8.0,
380
+ "eval_accuracy": 0.7476826783114993,
381
+ "eval_loss": 0.4009522497653961,
382
+ "eval_runtime": 15.0512,
383
+ "eval_samples_per_second": 33.22,
384
+ "eval_steps_per_second": 2.126,
385
+ "step": 1984
386
+ },
387
+ {
388
+ "epoch": 8.0,
389
+ "eval_exact_match": 17.4,
390
+ "eval_f1": 24.613044733044738,
391
+ "eval_qa_bleu": 11.328428780013391,
392
+ "eval_qa_exact_match": 0.162,
393
+ "eval_recite_bleu": 45.62279446830165,
394
+ "eval_recite_exact_match": 0.036,
395
+ "step": 1984
396
+ },
397
+ {
398
+ "epoch": 8.06,
399
+ "learning_rate": 3e-05,
400
+ "loss": 0.1383,
401
+ "step": 2000
402
+ },
403
+ {
404
+ "epoch": 8.27,
405
+ "learning_rate": 3e-05,
406
+ "loss": 0.1181,
407
+ "step": 2050
408
+ },
409
+ {
410
+ "epoch": 8.47,
411
+ "learning_rate": 3e-05,
412
+ "loss": 0.1212,
413
+ "step": 2100
414
+ },
415
+ {
416
+ "epoch": 8.67,
417
+ "learning_rate": 3e-05,
418
+ "loss": 0.118,
419
+ "step": 2150
420
+ },
421
+ {
422
+ "epoch": 8.87,
423
+ "learning_rate": 3e-05,
424
+ "loss": 0.1264,
425
+ "step": 2200
426
+ },
427
+ {
428
+ "epoch": 9.0,
429
+ "eval_accuracy": 0.7484366812227075,
430
+ "eval_loss": 0.39788317680358887,
431
+ "eval_runtime": 15.8038,
432
+ "eval_samples_per_second": 31.638,
433
+ "eval_steps_per_second": 2.025,
434
+ "step": 2232
435
+ },
436
+ {
437
+ "epoch": 9.0,
438
+ "eval_exact_match": 16.4,
439
+ "eval_f1": 23.748124098124105,
440
+ "eval_qa_bleu": 10.821376307326117,
441
+ "eval_qa_exact_match": 0.148,
442
+ "eval_recite_bleu": 46.63269966666572,
443
+ "eval_recite_exact_match": 0.04,
444
+ "step": 2232
445
+ },
446
+ {
447
+ "epoch": 9.07,
448
+ "learning_rate": 3e-05,
449
+ "loss": 0.121,
450
+ "step": 2250
451
+ },
452
+ {
453
+ "epoch": 9.27,
454
+ "learning_rate": 3e-05,
455
+ "loss": 0.1036,
456
+ "step": 2300
457
+ },
458
+ {
459
+ "epoch": 9.48,
460
+ "learning_rate": 3e-05,
461
+ "loss": 0.1053,
462
+ "step": 2350
463
+ },
464
+ {
465
+ "epoch": 9.68,
466
+ "learning_rate": 3e-05,
467
+ "loss": 0.1129,
468
+ "step": 2400
469
+ },
470
+ {
471
+ "epoch": 9.88,
472
+ "learning_rate": 3e-05,
473
+ "loss": 0.1091,
474
+ "step": 2450
475
+ },
476
+ {
477
+ "epoch": 10.0,
478
+ "eval_accuracy": 0.7487336244541485,
479
+ "eval_loss": 0.4059823751449585,
480
+ "eval_runtime": 15.4716,
481
+ "eval_samples_per_second": 32.317,
482
+ "eval_steps_per_second": 2.068,
483
+ "step": 2480
484
+ },
485
+ {
486
+ "epoch": 10.0,
487
+ "eval_exact_match": 16.6,
488
+ "eval_f1": 24.02516483516484,
489
+ "eval_qa_bleu": 9.045796919947849,
490
+ "eval_qa_exact_match": 0.156,
491
+ "eval_recite_bleu": 50.48708434203323,
492
+ "eval_recite_exact_match": 0.042,
493
+ "step": 2480
494
+ }
495
+ ],
496
+ "logging_steps": 50,
497
+ "max_steps": 4960,
498
+ "num_train_epochs": 20,
499
+ "save_steps": 500,
500
+ "total_flos": 1.152564234928128e+17,
501
+ "trial_name": null,
502
+ "trial_params": null
503
+ }
checkpoint-2480/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:acdd81e53b27f902dc4c1896d0c5253609dec7800ac5e5d60f5a323620c45c4a
3
+ size 4728
checkpoint-2480/vocab.json ADDED
The diff for this file is too large to render. See raw diff