Hugging Face's logo

tyzhu
/
lmind_nq_train6000_eval6489_v1_qa_5e-4_lora2

Safetensors
Generated from Trainer
Eval Results
Model card Files Community
tyzhu commited on
Commit
4fc3d2f
·
verified ·
1 Parent(s): 70c4969

Training in progress, epoch 16, checkpoint

Browse files
Files changed (14) hide show
  1. checkpoint-3187/adapter_config.json +1 -1
  2. checkpoint-3187/adapter_model.bin +3 -0
  3. checkpoint-3187/added_tokens.json +3 -3
  4. checkpoint-3187/optimizer.pt +2 -2
  5. checkpoint-3187/rng_state_0.pth +1 -1
  6. checkpoint-3187/rng_state_1.pth +1 -1
  7. checkpoint-3187/rng_state_2.pth +1 -1
  8. checkpoint-3187/rng_state_3.pth +1 -1
  9. checkpoint-3187/special_tokens_map.json +4 -12
  10. checkpoint-3187/tokenizer.json +0 -0
  11. checkpoint-3187/tokenizer.model +3 -0
  12. checkpoint-3187/tokenizer_config.json +17 -20
  13. checkpoint-3187/trainer_state.json +190 -235
  14. checkpoint-3187/training_args.bin +2 -2
checkpoint-3187/adapter_config.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "auto_mapping": null,
3
- "base_model_name_or_path": "Qwen/Qwen1.5-4B",
4
  "bias": "none",
5
  "fan_in_fan_out": false,
6
  "inference_mode": true,
 
1
  {
2
  "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
4
  "bias": "none",
5
  "fan_in_fan_out": false,
6
  "inference_mode": true,
checkpoint-3187/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3fcab80ae145c4bbb9c4481e97a99ae8804df36520c2cb69ed588bcc0dfd1cb9
3
+ size 143269386
checkpoint-3187/added_tokens.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "<|endoftext|>": 151643,
3
- "<|im_end|>": 151645,
4
- "<|im_start|>": 151644
5
  }
 
1
  {
2
+ "</s>": 2,
3
+ "<s>": 1,
4
+ "<unk>": 0
5
  }
checkpoint-3187/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9866cc192e1f7919c2566bf8ae62dc309e43e401c548fb6fba1c21225c7f184a
3
- size 224537202
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc13b5d7a707a4ee0fcb7b01cf25eb1975a8e4ef7238ada4238361b89ae0f50b
3
+ size 286585234
checkpoint-3187/rng_state_0.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:59b82c3b3be9e169e509b81f7aba5c3159f3d93cc92e5394f725a8008c958ce4
3
  size 15024
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8036dd1075ab81ed4af2b0ab7c5fb0a35da20ed9b706687cb7d7a8b72636ae85
3
  size 15024
checkpoint-3187/rng_state_1.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f92a850ca1a61a26db4df56f7559e816d78c151286597b074a40c80703f5f56b
3
  size 15024
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e2ed6fcc73b91e4be510632feea9881748aeff5c5d1936ad2ad4f94e0f059aa
3
  size 15024
checkpoint-3187/rng_state_2.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e4488bd8dfe351180ca749325bc2cf235952db8bba4a647e69c4383a98a84426
3
  size 15024
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:203989f1c2e1530b6bea85c2044e6dceafd34cf5aa0e8d5c04cfc13573c32140
3
  size 15024
checkpoint-3187/rng_state_3.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5b5ea62a0801144e3118bc1772f539e951799c769d5f7bb8ba7a3f04d485ff37
3
  size 15024
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f94ae6a06f223d8b767507241712d0effc4af664affa7a78f7b6f97a3ebfa2c4
3
  size 15024
checkpoint-3187/special_tokens_map.json CHANGED
@@ -1,14 +1,6 @@
1
  {
2
- "additional_special_tokens": [
3
- "<|im_start|>",
4
- "<|im_end|>"
5
- ],
6
- "eos_token": {
7
- "content": "<|endoftext|>",
8
- "lstrip": false,
9
- "normalized": false,
10
- "rstrip": false,
11
- "single_word": false
12
- },
13
- "pad_token": "<|endoftext|>"
14
  }
 
1
  {
2
+ "bos_token": "<s>",
3
+ "eos_token": "</s>",
4
+ "pad_token": "</s>",
5
+ "unk_token": "<unk>"
 
 
 
 
 
 
 
 
6
  }
checkpoint-3187/tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff
 
checkpoint-3187/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
checkpoint-3187/tokenizer_config.json CHANGED
@@ -1,24 +1,23 @@
1
  {
2
- "add_prefix_space": false,
3
  "added_tokens_decoder": {
4
- "151643": {
5
- "content": "<|endoftext|>",
6
  "lstrip": false,
7
  "normalized": false,
8
  "rstrip": false,
9
  "single_word": false,
10
  "special": true
11
  },
12
- "151644": {
13
- "content": "<|im_start|>",
14
  "lstrip": false,
15
  "normalized": false,
16
  "rstrip": false,
17
  "single_word": false,
18
  "special": true
19
  },
20
- "151645": {
21
- "content": "<|im_end|>",
22
  "lstrip": false,
23
  "normalized": false,
24
  "rstrip": false,
@@ -26,18 +25,16 @@
26
  "special": true
27
  }
28
  },
29
- "additional_special_tokens": [
30
- "<|im_start|>",
31
- "<|im_end|>"
32
- ],
33
- "bos_token": null,
34
- "chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
35
  "clean_up_tokenization_spaces": false,
36
- "eos_token": "<|endoftext|>",
37
- "errors": "replace",
38
- "model_max_length": 32768,
39
- "pad_token": "<|endoftext|>",
40
- "split_special_tokens": false,
41
- "tokenizer_class": "Qwen2Tokenizer",
42
- "unk_token": null
 
 
43
  }
 
1
  {
 
2
  "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<unk>",
5
  "lstrip": false,
6
  "normalized": false,
7
  "rstrip": false,
8
  "single_word": false,
9
  "special": true
10
  },
11
+ "1": {
12
+ "content": "<s>",
13
  "lstrip": false,
14
  "normalized": false,
15
  "rstrip": false,
16
  "single_word": false,
17
  "special": true
18
  },
19
+ "2": {
20
+ "content": "</s>",
21
  "lstrip": false,
22
  "normalized": false,
23
  "rstrip": false,
 
25
  "special": true
26
  }
27
  },
28
+ "additional_special_tokens": [],
29
+ "bos_token": "<s>",
 
 
 
 
30
  "clean_up_tokenization_spaces": false,
31
+ "eos_token": "</s>",
32
+ "legacy": false,
33
+ "model_max_length": 1000000000000000019884624838656,
34
+ "pad_token": "</s>",
35
+ "padding_side": "left",
36
+ "sp_model_kwargs": {},
37
+ "tokenizer_class": "LlamaTokenizer",
38
+ "unk_token": "<unk>",
39
+ "use_default_system_prompt": true
40
  }
checkpoint-3187/trainer_state.json CHANGED
@@ -9,497 +9,452 @@
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
- "epoch": 0.5333333333333333,
13
- "grad_norm": 0.7527768015861511,
14
  "learning_rate": 0.0005,
15
- "loss": 1.7217,
16
  "step": 100
17
  },
18
  {
19
- "epoch": 0.9973333333333333,
20
- "eval_accuracy": 0.5738974358974359,
21
- "eval_loss": 1.6209890842437744,
22
- "eval_runtime": 5.1723,
23
- "eval_samples_per_second": 96.668,
24
- "eval_steps_per_second": 12.18,
25
  "step": 187
26
  },
27
  {
28
- "epoch": 0.9973333333333333,
29
- "eval_exact_match": 18.2,
30
- "eval_f1": 27.000476190476196,
31
  "step": 187
32
  },
33
  {
34
- "epoch": 1.0666666666666667,
35
- "grad_norm": 0.7723727226257324,
36
  "learning_rate": 0.0005,
37
- "loss": 1.615,
38
  "step": 200
39
  },
40
  {
41
  "epoch": 1.6,
42
- "grad_norm": 0.8481244444847107,
43
  "learning_rate": 0.0005,
44
- "loss": 1.2605,
45
  "step": 300
46
  },
47
  {
48
  "epoch": 2.0,
49
- "eval_accuracy": 0.5706666666666667,
50
- "eval_loss": 1.6909353733062744,
51
- "eval_runtime": 5.4194,
52
- "eval_samples_per_second": 92.26,
53
- "eval_steps_per_second": 11.625,
54
  "step": 375
55
  },
56
  {
57
  "epoch": 2.0,
58
- "eval_exact_match": 17.0,
59
- "eval_f1": 25.9057142857143,
60
  "step": 375
61
  },
62
  {
63
- "epoch": 2.1333333333333333,
64
- "grad_norm": 0.810580313205719,
65
  "learning_rate": 0.0005,
66
- "loss": 1.1991,
67
  "step": 400
68
  },
69
  {
70
- "epoch": 2.6666666666666665,
71
- "grad_norm": 0.8971414566040039,
72
  "learning_rate": 0.0005,
73
- "loss": 0.9141,
74
  "step": 500
75
  },
76
  {
77
- "epoch": 2.997333333333333,
78
- "eval_accuracy": 0.5657435897435897,
79
- "eval_loss": 1.831854224205017,
80
- "eval_runtime": 4.9933,
81
- "eval_samples_per_second": 100.134,
82
- "eval_steps_per_second": 12.617,
83
  "step": 562
84
  },
85
  {
86
- "epoch": 2.997333333333333,
87
- "eval_exact_match": 16.0,
88
- "eval_f1": 23.940000000000012,
89
  "step": 562
90
  },
91
  {
92
  "epoch": 3.2,
93
- "grad_norm": 0.7627168297767639,
94
  "learning_rate": 0.0005,
95
- "loss": 0.8634,
96
  "step": 600
97
  },
98
  {
99
- "epoch": 3.7333333333333334,
100
- "grad_norm": 0.7384397387504578,
101
  "learning_rate": 0.0005,
102
- "loss": 0.7223,
103
  "step": 700
104
  },
105
  {
106
  "epoch": 4.0,
107
- "eval_accuracy": 0.5637948717948718,
108
- "eval_loss": 1.9418420791625977,
109
- "eval_runtime": 5.3476,
110
- "eval_samples_per_second": 93.499,
111
- "eval_steps_per_second": 11.781,
112
  "step": 750
113
  },
114
  {
115
  "epoch": 4.0,
116
- "eval_exact_match": 16.6,
117
- "eval_f1": 24.14476190476191,
118
  "step": 750
119
  },
120
  {
121
- "epoch": 4.266666666666667,
122
- "grad_norm": 0.7065069079399109,
123
  "learning_rate": 0.0005,
124
- "loss": 0.6786,
125
  "step": 800
126
  },
127
  {
128
  "epoch": 4.8,
129
- "grad_norm": 0.7625908851623535,
130
  "learning_rate": 0.0005,
131
- "loss": 0.6434,
132
  "step": 900
133
  },
134
  {
135
- "epoch": 4.997333333333334,
136
- "eval_accuracy": 0.5602051282051282,
137
- "eval_loss": 2.093710422515869,
138
- "eval_runtime": 5.1796,
139
- "eval_samples_per_second": 96.532,
140
- "eval_steps_per_second": 12.163,
141
  "step": 937
142
  },
143
  {
144
- "epoch": 4.997333333333334,
145
- "eval_exact_match": 17.2,
146
- "eval_f1": 24.822857142857156,
147
  "step": 937
148
  },
149
  {
150
- "epoch": 5.333333333333333,
151
- "grad_norm": 0.6257635951042175,
152
  "learning_rate": 0.0005,
153
- "loss": 0.6022,
154
  "step": 1000
155
  },
156
  {
157
- "epoch": 5.866666666666667,
158
- "grad_norm": 0.7330732345581055,
159
  "learning_rate": 0.0005,
160
- "loss": 0.5957,
161
  "step": 1100
162
  },
163
  {
164
  "epoch": 6.0,
165
- "eval_accuracy": 0.5573333333333333,
166
- "eval_loss": 2.17874813079834,
167
- "eval_runtime": 5.3048,
168
- "eval_samples_per_second": 94.255,
169
- "eval_steps_per_second": 11.876,
170
  "step": 1125
171
  },
172
  {
173
  "epoch": 6.0,
174
- "eval_exact_match": 15.2,
175
- "eval_f1": 22.341269841269852,
176
  "step": 1125
177
  },
178
  {
179
  "epoch": 6.4,
180
- "grad_norm": 0.5348079204559326,
181
  "learning_rate": 0.0005,
182
- "loss": 0.5583,
183
  "step": 1200
184
  },
185
  {
186
- "epoch": 6.933333333333334,
187
- "grad_norm": 0.4842449426651001,
188
  "learning_rate": 0.0005,
189
- "loss": 0.5728,
190
  "step": 1300
191
  },
192
  {
193
- "epoch": 6.997333333333334,
194
- "eval_accuracy": 0.558,
195
- "eval_loss": 2.2602250576019287,
196
- "eval_runtime": 5.068,
197
- "eval_samples_per_second": 98.658,
198
- "eval_steps_per_second": 12.431,
199
  "step": 1312
200
  },
201
  {
202
- "epoch": 6.997333333333334,
203
- "eval_exact_match": 15.8,
204
- "eval_f1": 22.874285714285733,
205
  "step": 1312
206
  },
207
  {
208
- "epoch": 7.466666666666667,
209
- "grad_norm": 0.5476555824279785,
210
  "learning_rate": 0.0005,
211
- "loss": 0.5339,
212
  "step": 1400
213
  },
214
  {
215
  "epoch": 8.0,
216
- "grad_norm": 0.8545846939086914,
217
  "learning_rate": 0.0005,
218
- "loss": 0.5596,
219
  "step": 1500
220
  },
221
  {
222
  "epoch": 8.0,
223
- "eval_accuracy": 0.5559487179487179,
224
- "eval_loss": 2.2222671508789062,
225
- "eval_runtime": 5.0471,
226
- "eval_samples_per_second": 99.067,
227
- "eval_steps_per_second": 12.482,
228
  "step": 1500
229
  },
230
  {
231
  "epoch": 8.0,
232
- "eval_exact_match": 15.2,
233
- "eval_f1": 23.320000000000007,
234
  "step": 1500
235
  },
236
  {
237
- "epoch": 8.533333333333333,
238
- "grad_norm": 0.5809034705162048,
239
  "learning_rate": 0.0005,
240
- "loss": 0.5218,
241
  "step": 1600
242
  },
243
  {
244
- "epoch": 8.997333333333334,
245
- "eval_accuracy": 0.5538461538461539,
246
- "eval_loss": 2.2796590328216553,
247
- "eval_runtime": 5.0384,
248
- "eval_samples_per_second": 99.238,
249
- "eval_steps_per_second": 12.504,
250
  "step": 1687
251
  },
252
  {
253
- "epoch": 8.997333333333334,
254
- "eval_exact_match": 13.2,
255
- "eval_f1": 20.368109668109675,
256
  "step": 1687
257
  },
258
  {
259
- "epoch": 9.066666666666666,
260
- "grad_norm": 0.48757097125053406,
261
  "learning_rate": 0.0005,
262
- "loss": 0.5444,
263
  "step": 1700
264
  },
265
  {
266
  "epoch": 9.6,
267
- "grad_norm": 0.5225056409835815,
268
  "learning_rate": 0.0005,
269
- "loss": 0.5254,
270
  "step": 1800
271
  },
272
  {
273
  "epoch": 10.0,
274
- "eval_accuracy": 0.5561025641025641,
275
- "eval_loss": 2.275782823562622,
276
- "eval_runtime": 5.158,
277
- "eval_samples_per_second": 96.937,
278
- "eval_steps_per_second": 12.214,
279
  "step": 1875
280
  },
281
  {
282
  "epoch": 10.0,
283
- "eval_exact_match": 14.0,
284
- "eval_f1": 21.65190476190477,
285
  "step": 1875
286
  },
287
  {
288
- "epoch": 10.133333333333333,
289
- "grad_norm": 0.7123950719833374,
290
  "learning_rate": 0.0005,
291
- "loss": 0.5304,
292
  "step": 1900
293
  },
294
  {
295
- "epoch": 10.666666666666666,
296
- "grad_norm": 0.48232755064964294,
297
  "learning_rate": 0.0005,
298
- "loss": 0.526,
299
  "step": 2000
300
  },
301
  {
302
- "epoch": 10.997333333333334,
303
- "eval_accuracy": 0.5544102564102564,
304
- "eval_loss": 2.2686431407928467,
305
- "eval_runtime": 5.0517,
306
- "eval_samples_per_second": 98.976,
307
- "eval_steps_per_second": 12.471,
308
  "step": 2062
309
  },
310
  {
311
- "epoch": 10.997333333333334,
312
- "eval_exact_match": 13.0,
313
- "eval_f1": 20.387460317460324,
314
  "step": 2062
315
  },
316
  {
317
  "epoch": 11.2,
318
- "grad_norm": 0.6355456113815308,
319
  "learning_rate": 0.0005,
320
- "loss": 0.5335,
321
  "step": 2100
322
  },
323
  {
324
- "epoch": 11.733333333333333,
325
- "grad_norm": 0.7310680150985718,
326
  "learning_rate": 0.0005,
327
- "loss": 0.5396,
328
  "step": 2200
329
  },
330
  {
331
  "epoch": 12.0,
332
- "eval_accuracy": 0.5533846153846154,
333
- "eval_loss": 2.2792105674743652,
334
- "eval_runtime": 5.1276,
335
- "eval_samples_per_second": 97.511,
336
- "eval_steps_per_second": 12.286,
337
  "step": 2250
338
  },
339
  {
340
  "epoch": 12.0,
341
- "eval_exact_match": 13.4,
342
- "eval_f1": 20.533650793650807,
343
  "step": 2250
344
  },
345
  {
346
- "epoch": 12.266666666666667,
347
- "grad_norm": 0.8953878879547119,
348
  "learning_rate": 0.0005,
349
- "loss": 0.5383,
350
  "step": 2300
351
  },
352
  {
353
  "epoch": 12.8,
354
- "grad_norm": 0.8268033266067505,
355
  "learning_rate": 0.0005,
356
- "loss": 0.5466,
357
  "step": 2400
358
  },
359
  {
360
- "epoch": 12.997333333333334,
361
- "eval_accuracy": 0.5500512820512821,
362
- "eval_loss": 2.2915709018707275,
363
- "eval_runtime": 5.0459,
364
- "eval_samples_per_second": 99.089,
365
- "eval_steps_per_second": 12.485,
366
  "step": 2437
367
  },
368
  {
369
- "epoch": 12.997333333333334,
370
- "eval_exact_match": 14.0,
371
- "eval_f1": 21.203650793650798,
372
  "step": 2437
373
  },
374
  {
375
- "epoch": 13.333333333333334,
376
- "grad_norm": 0.9471064209938049,
377
  "learning_rate": 0.0005,
378
- "loss": 0.5417,
379
  "step": 2500
380
  },
381
  {
382
- "epoch": 13.866666666666667,
383
- "grad_norm": 1.0178123712539673,
384
  "learning_rate": 0.0005,
385
- "loss": 0.556,
386
  "step": 2600
387
  },
388
  {
389
  "epoch": 14.0,
390
- "eval_accuracy": 0.5496923076923077,
391
- "eval_loss": 2.3209521770477295,
392
- "eval_runtime": 5.4611,
393
- "eval_samples_per_second": 91.557,
394
- "eval_steps_per_second": 11.536,
395
  "step": 2625
396
  },
397
  {
398
  "epoch": 14.0,
399
- "eval_exact_match": 12.6,
400
- "eval_f1": 20.174790764790767,
401
  "step": 2625
402
  },
403
  {
404
  "epoch": 14.4,
405
- "grad_norm": 0.750542402267456,
406
  "learning_rate": 0.0005,
407
- "loss": 0.5322,
408
  "step": 2700
409
  },
410
  {
411
- "epoch": 14.933333333333334,
412
- "grad_norm": 0.8345568776130676,
413
  "learning_rate": 0.0005,
414
- "loss": 0.5501,
415
  "step": 2800
416
  },
417
  {
418
- "epoch": 14.997333333333334,
419
- "eval_accuracy": 0.5464615384615384,
420
- "eval_loss": 2.3918418884277344,
421
- "eval_runtime": 5.1243,
422
- "eval_samples_per_second": 97.574,
423
- "eval_steps_per_second": 12.294,
424
  "step": 2812
425
  },
426
  {
427
- "epoch": 14.997333333333334,
428
- "eval_exact_match": 12.4,
429
- "eval_f1": 19.851428571428578,
430
  "step": 2812
431
  },
432
  {
433
- "epoch": 15.466666666666667,
434
- "grad_norm": 0.60535728931427,
435
  "learning_rate": 0.0005,
436
- "loss": 0.5174,
437
  "step": 2900
438
  },
439
  {
440
  "epoch": 16.0,
441
- "grad_norm": 0.7003186941146851,
442
  "learning_rate": 0.0005,
443
- "loss": 0.5461,
444
  "step": 3000
445
  },
446
  {
447
  "epoch": 16.0,
448
- "eval_accuracy": 0.5475897435897435,
449
- "eval_loss": 2.3702445030212402,
450
- "eval_runtime": 5.6883,
451
- "eval_samples_per_second": 87.9,
452
- "eval_steps_per_second": 11.075,
453
  "step": 3000
454
  },
455
  {
456
  "epoch": 16.0,
457
- "eval_exact_match": 12.6,
458
- "eval_f1": 19.352380952380955,
459
  "step": 3000
460
  },
461
  {
462
- "epoch": 16.533333333333335,
463
- "grad_norm": 0.9538315534591675,
464
  "learning_rate": 0.0005,
465
- "loss": 0.5032,
466
  "step": 3100
467
  },
468
  {
469
- "epoch": 16.997333333333334,
470
- "eval_accuracy": 0.5461025641025641,
471
- "eval_loss": 2.443984270095825,
472
- "eval_runtime": 5.1587,
473
- "eval_samples_per_second": 96.923,
474
- "eval_steps_per_second": 12.212,
475
  "step": 3187
476
  },
477
  {
478
- "epoch": 16.997333333333334,
479
- "eval_exact_match": 11.6,
480
- "eval_f1": 19.255238095238102,
481
  "step": 3187
482
  }
483
  ],
484
  "logging_steps": 100,
485
  "max_steps": 9350,
486
- "num_input_tokens_seen": 0,
487
  "num_train_epochs": 50,
488
  "save_steps": 500,
489
- "stateful_callbacks": {
490
- "TrainerControl": {
491
- "args": {
492
- "should_epoch_stop": false,
493
- "should_evaluate": false,
494
- "should_log": false,
495
- "should_save": true,
496
- "should_training_stop": false
497
- },
498
- "attributes": {}
499
- }
500
- },
501
- "total_flos": 5.551285710146765e+16,
502
- "train_batch_size": 1,
503
  "trial_name": null,
504
  "trial_params": null
505
  }
 
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
+ "epoch": 0.53,
 
13
  "learning_rate": 0.0005,
14
+ "loss": 1.43,
15
  "step": 100
16
  },
17
  {
18
+ "epoch": 1.0,
19
+ "eval_accuracy": 0.6161538461538462,
20
+ "eval_loss": 1.2682808637619019,
21
+ "eval_runtime": 4.85,
22
+ "eval_samples_per_second": 103.093,
23
+ "eval_steps_per_second": 12.99,
24
  "step": 187
25
  },
26
  {
27
+ "epoch": 1.0,
28
+ "eval_exact_match": 26.6,
29
+ "eval_f1": 38.8657142857143,
30
  "step": 187
31
  },
32
  {
33
+ "epoch": 1.07,
 
34
  "learning_rate": 0.0005,
35
+ "loss": 1.2637,
36
  "step": 200
37
  },
38
  {
39
  "epoch": 1.6,
 
40
  "learning_rate": 0.0005,
41
+ "loss": 1.0285,
42
  "step": 300
43
  },
44
  {
45
  "epoch": 2.0,
46
+ "eval_accuracy": 0.6129230769230769,
47
+ "eval_loss": 1.3220478296279907,
48
+ "eval_runtime": 4.4227,
49
+ "eval_samples_per_second": 113.052,
50
+ "eval_steps_per_second": 14.245,
51
  "step": 375
52
  },
53
  {
54
  "epoch": 2.0,
55
+ "eval_exact_match": 24.2,
56
+ "eval_f1": 37.12603174603175,
57
  "step": 375
58
  },
59
  {
60
+ "epoch": 2.13,
 
61
  "learning_rate": 0.0005,
62
+ "loss": 0.9478,
63
  "step": 400
64
  },
65
  {
66
+ "epoch": 2.67,
 
67
  "learning_rate": 0.0005,
68
+ "loss": 0.7318,
69
  "step": 500
70
  },
71
  {
72
+ "epoch": 3.0,
73
+ "eval_accuracy": 0.6075897435897436,
74
+ "eval_loss": 1.4644758701324463,
75
+ "eval_runtime": 4.8316,
76
+ "eval_samples_per_second": 103.486,
77
+ "eval_steps_per_second": 13.039,
78
  "step": 562
79
  },
80
  {
81
+ "epoch": 3.0,
82
+ "eval_exact_match": 22.8,
83
+ "eval_f1": 36.51031746031747,
84
  "step": 562
85
  },
86
  {
87
  "epoch": 3.2,
 
88
  "learning_rate": 0.0005,
89
+ "loss": 0.6922,
90
  "step": 600
91
  },
92
  {
93
+ "epoch": 3.73,
 
94
  "learning_rate": 0.0005,
95
+ "loss": 0.5898,
96
  "step": 700
97
  },
98
  {
99
  "epoch": 4.0,
100
+ "eval_accuracy": 0.604974358974359,
101
+ "eval_loss": 1.54537832736969,
102
+ "eval_runtime": 4.119,
103
+ "eval_samples_per_second": 121.388,
104
+ "eval_steps_per_second": 15.295,
105
  "step": 750
106
  },
107
  {
108
  "epoch": 4.0,
109
+ "eval_exact_match": 15.4,
110
+ "eval_f1": 25.034444444444457,
111
  "step": 750
112
  },
113
  {
114
+ "epoch": 4.27,
 
115
  "learning_rate": 0.0005,
116
+ "loss": 0.5599,
117
  "step": 800
118
  },
119
  {
120
  "epoch": 4.8,
 
121
  "learning_rate": 0.0005,
122
+ "loss": 0.5309,
123
  "step": 900
124
  },
125
  {
126
+ "epoch": 5.0,
127
+ "eval_accuracy": 0.6025641025641025,
128
+ "eval_loss": 1.6439025402069092,
129
+ "eval_runtime": 4.3353,
130
+ "eval_samples_per_second": 115.332,
131
+ "eval_steps_per_second": 14.532,
132
  "step": 937
133
  },
134
  {
135
+ "epoch": 5.0,
136
+ "eval_exact_match": 21.6,
137
+ "eval_f1": 34.31112554112555,
138
  "step": 937
139
  },
140
  {
141
+ "epoch": 5.33,
 
142
  "learning_rate": 0.0005,
143
+ "loss": 0.5011,
144
  "step": 1000
145
  },
146
  {
147
+ "epoch": 5.87,
 
148
  "learning_rate": 0.0005,
149
+ "loss": 0.4985,
150
  "step": 1100
151
  },
152
  {
153
  "epoch": 6.0,
154
+ "eval_accuracy": 0.6034358974358974,
155
+ "eval_loss": 1.722017526626587,
156
+ "eval_runtime": 4.2182,
157
+ "eval_samples_per_second": 118.533,
158
+ "eval_steps_per_second": 14.935,
159
  "step": 1125
160
  },
161
  {
162
  "epoch": 6.0,
163
+ "eval_exact_match": 22.4,
164
+ "eval_f1": 34.31126984126985,
165
  "step": 1125
166
  },
167
  {
168
  "epoch": 6.4,
 
169
  "learning_rate": 0.0005,
170
+ "loss": 0.4686,
171
  "step": 1200
172
  },
173
  {
174
+ "epoch": 6.93,
 
175
  "learning_rate": 0.0005,
176
+ "loss": 0.5091,
177
  "step": 1300
178
  },
179
  {
180
+ "epoch": 7.0,
181
+ "eval_accuracy": 0.6008205128205129,
182
+ "eval_loss": 1.800759196281433,
183
+ "eval_runtime": 4.8504,
184
+ "eval_samples_per_second": 103.083,
185
+ "eval_steps_per_second": 12.989,
186
  "step": 1312
187
  },
188
  {
189
+ "epoch": 7.0,
190
+ "eval_exact_match": 0.2,
191
+ "eval_f1": 2.1497326649958226,
192
  "step": 1312
193
  },
194
  {
195
+ "epoch": 7.47,
 
196
  "learning_rate": 0.0005,
197
+ "loss": 0.4576,
198
  "step": 1400
199
  },
200
  {
201
  "epoch": 8.0,
 
202
  "learning_rate": 0.0005,
203
+ "loss": 0.4796,
204
  "step": 1500
205
  },
206
  {
207
  "epoch": 8.0,
208
+ "eval_accuracy": 0.6001025641025641,
209
+ "eval_loss": 1.7781891822814941,
210
+ "eval_runtime": 4.4147,
211
+ "eval_samples_per_second": 113.257,
212
+ "eval_steps_per_second": 14.27,
213
  "step": 1500
214
  },
215
  {
216
  "epoch": 8.0,
217
+ "eval_exact_match": 19.2,
218
+ "eval_f1": 29.972380952380963,
219
  "step": 1500
220
  },
221
  {
222
+ "epoch": 8.53,
 
223
  "learning_rate": 0.0005,
224
+ "loss": 0.4453,
225
  "step": 1600
226
  },
227
  {
228
+ "epoch": 9.0,
229
+ "eval_accuracy": 0.5984615384615385,
230
+ "eval_loss": 1.8255126476287842,
231
+ "eval_runtime": 4.1057,
232
+ "eval_samples_per_second": 121.781,
233
+ "eval_steps_per_second": 15.344,
234
  "step": 1687
235
  },
236
  {
237
+ "epoch": 9.0,
238
+ "eval_exact_match": 17.8,
239
+ "eval_f1": 29.10082251082252,
240
  "step": 1687
241
  },
242
  {
243
+ "epoch": 9.07,
 
244
  "learning_rate": 0.0005,
245
+ "loss": 0.4701,
246
  "step": 1700
247
  },
248
  {
249
  "epoch": 9.6,
 
250
  "learning_rate": 0.0005,
251
+ "loss": 0.448,
252
  "step": 1800
253
  },
254
  {
255
  "epoch": 10.0,
256
+ "eval_accuracy": 0.5931282051282051,
257
+ "eval_loss": 1.7979023456573486,
258
+ "eval_runtime": 4.4695,
259
+ "eval_samples_per_second": 111.87,
260
+ "eval_steps_per_second": 14.096,
261
  "step": 1875
262
  },
263
  {
264
  "epoch": 10.0,
265
+ "eval_exact_match": 18.8,
266
+ "eval_f1": 28.28873015873017,
267
  "step": 1875
268
  },
269
  {
270
+ "epoch": 10.13,
 
271
  "learning_rate": 0.0005,
272
+ "loss": 0.4641,
273
  "step": 1900
274
  },
275
  {
276
+ "epoch": 10.67,
 
277
  "learning_rate": 0.0005,
278
+ "loss": 0.4522,
279
  "step": 2000
280
  },
281
  {
282
+ "epoch": 11.0,
283
+ "eval_accuracy": 0.595948717948718,
284
+ "eval_loss": 1.8272360563278198,
285
+ "eval_runtime": 4.1203,
286
+ "eval_samples_per_second": 121.35,
287
+ "eval_steps_per_second": 15.29,
288
  "step": 2062
289
  },
290
  {
291
+ "epoch": 11.0,
292
+ "eval_exact_match": 17.2,
293
+ "eval_f1": 28.030476190476207,
294
  "step": 2062
295
  },
296
  {
297
  "epoch": 11.2,
 
298
  "learning_rate": 0.0005,
299
+ "loss": 0.4555,
300
  "step": 2100
301
  },
302
  {
303
+ "epoch": 11.73,
 
304
  "learning_rate": 0.0005,
305
+ "loss": 0.4552,
306
  "step": 2200
307
  },
308
  {
309
  "epoch": 12.0,
310
+ "eval_accuracy": 0.5946153846153847,
311
+ "eval_loss": 1.8670085668563843,
312
+ "eval_runtime": 4.2129,
313
+ "eval_samples_per_second": 118.684,
314
+ "eval_steps_per_second": 14.954,
315
  "step": 2250
316
  },
317
  {
318
  "epoch": 12.0,
319
+ "eval_exact_match": 17.8,
320
+ "eval_f1": 28.45285714285715,
321
  "step": 2250
322
  },
323
  {
324
+ "epoch": 12.27,
 
325
  "learning_rate": 0.0005,
326
+ "loss": 0.4492,
327
  "step": 2300
328
  },
329
  {
330
  "epoch": 12.8,
 
331
  "learning_rate": 0.0005,
332
+ "loss": 0.4551,
333
  "step": 2400
334
  },
335
  {
336
+ "epoch": 13.0,
337
+ "eval_accuracy": 0.595025641025641,
338
+ "eval_loss": 1.8706276416778564,
339
+ "eval_runtime": 4.2102,
340
+ "eval_samples_per_second": 118.759,
341
+ "eval_steps_per_second": 14.964,
342
  "step": 2437
343
  },
344
  {
345
+ "epoch": 13.0,
346
+ "eval_exact_match": 19.0,
347
+ "eval_f1": 28.634603174603175,
348
  "step": 2437
349
  },
350
  {
351
+ "epoch": 13.33,
 
352
  "learning_rate": 0.0005,
353
+ "loss": 0.4505,
354
  "step": 2500
355
  },
356
  {
357
+ "epoch": 13.87,
 
358
  "learning_rate": 0.0005,
359
+ "loss": 0.4559,
360
  "step": 2600
361
  },
362
  {
363
  "epoch": 14.0,
364
+ "eval_accuracy": 0.5925128205128205,
365
+ "eval_loss": 1.8730883598327637,
366
+ "eval_runtime": 4.1209,
367
+ "eval_samples_per_second": 121.333,
368
+ "eval_steps_per_second": 15.288,
369
  "step": 2625
370
  },
371
  {
372
  "epoch": 14.0,
373
+ "eval_exact_match": 16.4,
374
+ "eval_f1": 25.989682539682548,
375
  "step": 2625
376
  },
377
  {
378
  "epoch": 14.4,
 
379
  "learning_rate": 0.0005,
380
+ "loss": 0.4455,
381
  "step": 2700
382
  },
383
  {
384
+ "epoch": 14.93,
 
385
  "learning_rate": 0.0005,
386
+ "loss": 0.4581,
387
  "step": 2800
388
  },
389
  {
390
+ "epoch": 15.0,
391
+ "eval_accuracy": 0.5932307692307692,
392
+ "eval_loss": 1.8531454801559448,
393
+ "eval_runtime": 4.4124,
394
+ "eval_samples_per_second": 113.316,
395
+ "eval_steps_per_second": 14.278,
396
  "step": 2812
397
  },
398
  {
399
+ "epoch": 15.0,
400
+ "eval_exact_match": 17.8,
401
+ "eval_f1": 28.33587301587303,
402
  "step": 2812
403
  },
404
  {
405
+ "epoch": 15.47,
 
406
  "learning_rate": 0.0005,
407
+ "loss": 0.4358,
408
  "step": 2900
409
  },
410
  {
411
  "epoch": 16.0,
 
412
  "learning_rate": 0.0005,
413
+ "loss": 0.4535,
414
  "step": 3000
415
  },
416
  {
417
  "epoch": 16.0,
418
+ "eval_accuracy": 0.5923076923076923,
419
+ "eval_loss": 1.949161410331726,
420
+ "eval_runtime": 4.8341,
421
+ "eval_samples_per_second": 103.433,
422
+ "eval_steps_per_second": 13.033,
423
  "step": 3000
424
  },
425
  {
426
  "epoch": 16.0,
427
+ "eval_exact_match": 17.4,
428
+ "eval_f1": 27.222698412698417,
429
  "step": 3000
430
  },
431
  {
432
+ "epoch": 16.53,
 
433
  "learning_rate": 0.0005,
434
+ "loss": 0.4308,
435
  "step": 3100
436
  },
437
  {
438
+ "epoch": 17.0,
439
+ "eval_accuracy": 0.5915384615384616,
440
+ "eval_loss": 1.8943943977355957,
441
+ "eval_runtime": 4.2634,
442
+ "eval_samples_per_second": 117.279,
443
+ "eval_steps_per_second": 14.777,
444
  "step": 3187
445
  },
446
  {
447
+ "epoch": 17.0,
448
+ "eval_exact_match": 17.4,
449
+ "eval_f1": 26.819047619047623,
450
  "step": 3187
451
  }
452
  ],
453
  "logging_steps": 100,
454
  "max_steps": 9350,
 
455
  "num_train_epochs": 50,
456
  "save_steps": 500,
457
+ "total_flos": 1.189519251514327e+17,
 
 
 
 
 
 
 
 
 
 
 
 
 
458
  "trial_name": null,
459
  "trial_params": null
460
  }
checkpoint-3187/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2776615b5656a131d464a15a3d5283d9bdf82687bc8ffc8d243c778d1158123d
3
- size 5304
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aeaafcc180d545c4c7bc83b3ef040967ab10c6394ea7a4c1dc01fb190e8fc740
3
+ size 4728