Hugging Face's logo

tyzhu
/
lmind_nq_train6000_eval6489_v1_qa_3e-4_lora2

Safetensors
Generated from Trainer
Eval Results
Model card Files Community
tyzhu commited on
Commit
9184d82
·
verified ·
1 Parent(s): 3919f19

Training in progress, epoch 12, checkpoint

Browse files
Files changed (14) hide show
  1. checkpoint-2437/adapter_config.json +1 -1
  2. checkpoint-2437/adapter_model.bin +3 -0
  3. checkpoint-2437/added_tokens.json +3 -3
  4. checkpoint-2437/optimizer.pt +2 -2
  5. checkpoint-2437/rng_state_0.pth +1 -1
  6. checkpoint-2437/rng_state_1.pth +1 -1
  7. checkpoint-2437/rng_state_2.pth +1 -1
  8. checkpoint-2437/rng_state_3.pth +1 -1
  9. checkpoint-2437/special_tokens_map.json +4 -12
  10. checkpoint-2437/tokenizer.json +0 -0
  11. checkpoint-2437/tokenizer.model +3 -0
  12. checkpoint-2437/tokenizer_config.json +17 -20
  13. checkpoint-2437/trainer_state.json +146 -184
  14. checkpoint-2437/training_args.bin +2 -2
checkpoint-2437/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-2437/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d9a2e6b277da55dba50e080d115a18c215d36b47e0b47fd5c94cf0b5efc9c3a
3
+ size 143269386
checkpoint-2437/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-2437/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e3ec7509baebabf1e0c1c2de8767e8db9d3ce6ca7e1bdb3798f39d014f94339b
3
- size 224537202
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a0b5863621b3496033377c01e354f0a1d6461e9fbfb2408dca26f96dc0520d0c
3
+ size 286585234
checkpoint-2437/rng_state_0.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:42ef19a59b0a7e99b66322574f07b85c83852e4e9b94ebcf250281ef42f38814
3
  size 15024
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1f62cfbecee25d844fc8be936984e69921f6551aff1848f2848d43ec56928b29
3
  size 15024
checkpoint-2437/rng_state_1.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:13a29e33b1c3467d2bf80bb112049dec1d7aa102a9661cafca3dcab43a04d947
3
  size 15024
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:52323c0b55c2e122711ce5c06f254b5a8bbfcf626506bcc9558f4c42d7af10d5
3
  size 15024
checkpoint-2437/rng_state_2.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:99782da560c289a64d6cdd477a104d15ef3f07d166e348f959cedc11bab46fc5
3
  size 15024
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a3e0bad4b81eaa806addcccff7106e177dd9345293214ba37c5bc9663991b314
3
  size 15024
checkpoint-2437/rng_state_3.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e89fb4a014d939b0cd4f7c1746b235d935bb14e96a018c6be879b50178b127e4
3
  size 15024
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5da30da00ffa7679d30f4389b18e4e3303b191ea3bd2541446817f21a1cda067
3
  size 15024
checkpoint-2437/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-2437/tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff
 
checkpoint-2437/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
checkpoint-2437/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-2437/trainer_state.json CHANGED
@@ -9,388 +9,350 @@
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
- "epoch": 0.5333333333333333,
13
- "grad_norm": 0.5194531679153442,
14
  "learning_rate": 0.0003,
15
- "loss": 1.7259,
16
  "step": 100
17
  },
18
  {
19
- "epoch": 0.9973333333333333,
20
- "eval_accuracy": 0.5753333333333334,
21
- "eval_loss": 1.6105104684829712,
22
- "eval_runtime": 5.418,
23
- "eval_samples_per_second": 92.285,
24
- "eval_steps_per_second": 11.628,
25
  "step": 187
26
  },
27
  {
28
- "epoch": 0.9973333333333333,
29
- "eval_exact_match": 17.2,
30
- "eval_f1": 26.890173160173163,
31
  "step": 187
32
  },
33
  {
34
- "epoch": 1.0666666666666667,
35
- "grad_norm": 0.6734262704849243,
36
  "learning_rate": 0.0003,
37
- "loss": 1.6156,
38
  "step": 200
39
  },
40
  {
41
  "epoch": 1.6,
42
- "grad_norm": 0.7422465682029724,
43
  "learning_rate": 0.0003,
44
- "loss": 1.3358,
45
  "step": 300
46
  },
47
  {
48
  "epoch": 2.0,
49
- "eval_accuracy": 0.5722564102564103,
50
- "eval_loss": 1.6464648246765137,
51
- "eval_runtime": 5.1856,
52
- "eval_samples_per_second": 96.422,
53
- "eval_steps_per_second": 12.149,
54
  "step": 375
55
  },
56
  {
57
  "epoch": 2.0,
58
- "eval_exact_match": 19.0,
59
- "eval_f1": 27.762380952380955,
60
  "step": 375
61
  },
62
  {
63
- "epoch": 2.1333333333333333,
64
- "grad_norm": 0.828364908695221,
65
  "learning_rate": 0.0003,
66
- "loss": 1.2607,
67
  "step": 400
68
  },
69
  {
70
- "epoch": 2.6666666666666665,
71
- "grad_norm": 0.9025559425354004,
72
  "learning_rate": 0.0003,
73
- "loss": 0.9854,
74
  "step": 500
75
  },
76
  {
77
- "epoch": 2.997333333333333,
78
- "eval_accuracy": 0.5707692307692308,
79
- "eval_loss": 1.7620763778686523,
80
- "eval_runtime": 5.2589,
81
- "eval_samples_per_second": 95.077,
82
- "eval_steps_per_second": 11.98,
83
  "step": 562
84
  },
85
  {
86
- "epoch": 2.997333333333333,
87
- "eval_exact_match": 18.2,
88
- "eval_f1": 27.53142857142857,
89
  "step": 562
90
  },
91
  {
92
  "epoch": 3.2,
93
- "grad_norm": 0.8217537999153137,
94
  "learning_rate": 0.0003,
95
- "loss": 0.9099,
96
  "step": 600
97
  },
98
  {
99
- "epoch": 3.7333333333333334,
100
- "grad_norm": 0.8315110206604004,
101
  "learning_rate": 0.0003,
102
- "loss": 0.7508,
103
  "step": 700
104
  },
105
  {
106
  "epoch": 4.0,
107
- "eval_accuracy": 0.5663076923076923,
108
- "eval_loss": 1.9380683898925781,
109
- "eval_runtime": 5.3338,
110
- "eval_samples_per_second": 93.742,
111
- "eval_steps_per_second": 11.811,
112
  "step": 750
113
  },
114
  {
115
  "epoch": 4.0,
116
- "eval_exact_match": 16.0,
117
- "eval_f1": 24.611904761904768,
118
  "step": 750
119
  },
120
  {
121
- "epoch": 4.266666666666667,
122
- "grad_norm": 0.6763857007026672,
123
  "learning_rate": 0.0003,
124
- "loss": 0.6912,
125
  "step": 800
126
  },
127
  {
128
  "epoch": 4.8,
129
- "grad_norm": 0.8408166766166687,
130
  "learning_rate": 0.0003,
131
- "loss": 0.6396,
132
  "step": 900
133
  },
134
  {
135
- "epoch": 4.997333333333334,
136
- "eval_accuracy": 0.5634358974358974,
137
- "eval_loss": 1.9925932884216309,
138
- "eval_runtime": 5.3717,
139
- "eval_samples_per_second": 93.08,
140
- "eval_steps_per_second": 11.728,
141
  "step": 937
142
  },
143
  {
144
- "epoch": 4.997333333333334,
145
- "eval_exact_match": 16.0,
146
- "eval_f1": 24.313809523809528,
147
  "step": 937
148
  },
149
  {
150
- "epoch": 5.333333333333333,
151
- "grad_norm": 0.6721271276473999,
152
  "learning_rate": 0.0003,
153
- "loss": 0.5959,
154
  "step": 1000
155
  },
156
  {
157
- "epoch": 5.866666666666667,
158
- "grad_norm": 0.9258321523666382,
159
  "learning_rate": 0.0003,
160
- "loss": 0.5833,
161
  "step": 1100
162
  },
163
  {
164
  "epoch": 6.0,
165
- "eval_accuracy": 0.5611794871794872,
166
- "eval_loss": 2.101524829864502,
167
- "eval_runtime": 5.4208,
168
- "eval_samples_per_second": 92.238,
169
- "eval_steps_per_second": 11.622,
170
  "step": 1125
171
  },
172
  {
173
  "epoch": 6.0,
174
- "eval_exact_match": 17.0,
175
- "eval_f1": 25.444285714285726,
176
  "step": 1125
177
  },
178
  {
179
  "epoch": 6.4,
180
- "grad_norm": 0.6029295921325684,
181
  "learning_rate": 0.0003,
182
- "loss": 0.545,
183
  "step": 1200
184
  },
185
  {
186
- "epoch": 6.933333333333334,
187
- "grad_norm": 0.5719736814498901,
188
  "learning_rate": 0.0003,
189
- "loss": 0.5567,
190
  "step": 1300
191
  },
192
  {
193
- "epoch": 6.997333333333334,
194
- "eval_accuracy": 0.5607179487179487,
195
- "eval_loss": 2.1645195484161377,
196
- "eval_runtime": 5.1974,
197
- "eval_samples_per_second": 96.202,
198
- "eval_steps_per_second": 12.121,
199
  "step": 1312
200
  },
201
  {
202
- "epoch": 6.997333333333334,
203
- "eval_exact_match": 15.0,
204
- "eval_f1": 23.618095238095247,
205
  "step": 1312
206
  },
207
  {
208
- "epoch": 7.466666666666667,
209
- "grad_norm": 0.49924588203430176,
210
  "learning_rate": 0.0003,
211
- "loss": 0.5205,
212
  "step": 1400
213
  },
214
  {
215
  "epoch": 8.0,
216
- "grad_norm": 0.5360875129699707,
217
  "learning_rate": 0.0003,
218
- "loss": 0.5411,
219
  "step": 1500
220
  },
221
  {
222
  "epoch": 8.0,
223
- "eval_accuracy": 0.5614358974358974,
224
- "eval_loss": 2.2039756774902344,
225
- "eval_runtime": 5.0936,
226
- "eval_samples_per_second": 98.162,
227
- "eval_steps_per_second": 12.368,
228
  "step": 1500
229
  },
230
  {
231
  "epoch": 8.0,
232
- "eval_exact_match": 16.8,
233
- "eval_f1": 25.132380952380966,
234
  "step": 1500
235
  },
236
  {
237
- "epoch": 8.533333333333333,
238
- "grad_norm": 0.44560280442237854,
239
  "learning_rate": 0.0003,
240
- "loss": 0.5028,
241
  "step": 1600
242
  },
243
  {
244
- "epoch": 8.997333333333334,
245
- "eval_accuracy": 0.5608205128205128,
246
- "eval_loss": 2.236473321914673,
247
- "eval_runtime": 5.2122,
248
- "eval_samples_per_second": 95.93,
249
- "eval_steps_per_second": 12.087,
250
  "step": 1687
251
  },
252
  {
253
- "epoch": 8.997333333333334,
254
- "eval_exact_match": 16.0,
255
- "eval_f1": 24.712380952380958,
256
  "step": 1687
257
  },
258
  {
259
- "epoch": 9.066666666666666,
260
- "grad_norm": 0.42847225069999695,
261
  "learning_rate": 0.0003,
262
- "loss": 0.5246,
263
  "step": 1700
264
  },
265
  {
266
  "epoch": 9.6,
267
- "grad_norm": 0.4186452627182007,
268
  "learning_rate": 0.0003,
269
- "loss": 0.5041,
270
  "step": 1800
271
  },
272
  {
273
  "epoch": 10.0,
274
- "eval_accuracy": 0.5604615384615385,
275
- "eval_loss": 2.2862467765808105,
276
- "eval_runtime": 5.8141,
277
- "eval_samples_per_second": 85.998,
278
- "eval_steps_per_second": 10.836,
279
  "step": 1875
280
  },
281
  {
282
  "epoch": 10.0,
283
- "eval_exact_match": 14.4,
284
- "eval_f1": 22.683333333333344,
285
  "step": 1875
286
  },
287
  {
288
- "epoch": 10.133333333333333,
289
- "grad_norm": 0.427721232175827,
290
  "learning_rate": 0.0003,
291
- "loss": 0.5083,
292
  "step": 1900
293
  },
294
  {
295
- "epoch": 10.666666666666666,
296
- "grad_norm": 0.4298701584339142,
297
  "learning_rate": 0.0003,
298
- "loss": 0.4991,
299
  "step": 2000
300
  },
301
  {
302
- "epoch": 10.997333333333334,
303
- "eval_accuracy": 0.5602564102564103,
304
- "eval_loss": 2.285149574279785,
305
- "eval_runtime": 5.1591,
306
- "eval_samples_per_second": 96.916,
307
- "eval_steps_per_second": 12.211,
308
  "step": 2062
309
  },
310
  {
311
- "epoch": 10.997333333333334,
312
- "eval_exact_match": 14.4,
313
- "eval_f1": 22.41619047619048,
314
  "step": 2062
315
  },
316
  {
317
  "epoch": 11.2,
318
- "grad_norm": 0.40195176005363464,
319
  "learning_rate": 0.0003,
320
- "loss": 0.5053,
321
  "step": 2100
322
  },
323
  {
324
- "epoch": 11.733333333333333,
325
- "grad_norm": 0.6030442118644714,
326
  "learning_rate": 0.0003,
327
- "loss": 0.5048,
328
  "step": 2200
329
  },
330
  {
331
  "epoch": 12.0,
332
- "eval_accuracy": 0.560974358974359,
333
- "eval_loss": 2.2454702854156494,
334
- "eval_runtime": 5.1149,
335
- "eval_samples_per_second": 97.753,
336
- "eval_steps_per_second": 12.317,
337
  "step": 2250
338
  },
339
  {
340
  "epoch": 12.0,
341
- "eval_exact_match": 16.2,
342
- "eval_f1": 24.622857142857157,
343
  "step": 2250
344
  },
345
  {
346
- "epoch": 12.266666666666667,
347
- "grad_norm": 0.4596252143383026,
348
  "learning_rate": 0.0003,
349
- "loss": 0.4995,
350
  "step": 2300
351
  },
352
  {
353
  "epoch": 12.8,
354
- "grad_norm": 0.4469389021396637,
355
  "learning_rate": 0.0003,
356
- "loss": 0.5067,
357
  "step": 2400
358
  },
359
  {
360
- "epoch": 12.997333333333334,
361
- "eval_accuracy": 0.5591794871794872,
362
- "eval_loss": 2.258878707885742,
363
- "eval_runtime": 5.1623,
364
- "eval_samples_per_second": 96.856,
365
- "eval_steps_per_second": 12.204,
366
  "step": 2437
367
  },
368
  {
369
- "epoch": 12.997333333333334,
370
- "eval_exact_match": 14.0,
371
- "eval_f1": 22.422539682539693,
372
  "step": 2437
373
  }
374
  ],
375
  "logging_steps": 100,
376
  "max_steps": 9350,
377
- "num_input_tokens_seen": 0,
378
  "num_train_epochs": 50,
379
  "save_steps": 500,
380
- "stateful_callbacks": {
381
- "TrainerControl": {
382
- "args": {
383
- "should_epoch_stop": false,
384
- "should_evaluate": false,
385
- "should_log": false,
386
- "should_save": true,
387
- "should_training_stop": false
388
- },
389
- "attributes": {}
390
- }
391
- },
392
- "total_flos": 4.245100839934362e+16,
393
- "train_batch_size": 1,
394
  "trial_name": null,
395
  "trial_params": null
396
  }
 
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
+ "epoch": 0.53,
 
13
  "learning_rate": 0.0003,
14
+ "loss": 1.4433,
15
  "step": 100
16
  },
17
  {
18
+ "epoch": 1.0,
19
+ "eval_accuracy": 0.6165641025641025,
20
+ "eval_loss": 1.2625622749328613,
21
+ "eval_runtime": 4.3923,
22
+ "eval_samples_per_second": 113.835,
23
+ "eval_steps_per_second": 14.343,
24
  "step": 187
25
  },
26
  {
27
+ "epoch": 1.0,
28
+ "eval_exact_match": 26.4,
29
+ "eval_f1": 39.52571428571429,
30
  "step": 187
31
  },
32
  {
33
+ "epoch": 1.07,
 
34
  "learning_rate": 0.0003,
35
+ "loss": 1.2659,
36
  "step": 200
37
  },
38
  {
39
  "epoch": 1.6,
 
40
  "learning_rate": 0.0003,
41
+ "loss": 1.0761,
42
  "step": 300
43
  },
44
  {
45
  "epoch": 2.0,
46
+ "eval_accuracy": 0.615897435897436,
47
+ "eval_loss": 1.295995831489563,
48
+ "eval_runtime": 4.4145,
49
+ "eval_samples_per_second": 113.264,
50
+ "eval_steps_per_second": 14.271,
51
  "step": 375
52
  },
53
  {
54
  "epoch": 2.0,
55
+ "eval_exact_match": 25.4,
56
+ "eval_f1": 38.36603174603176,
57
  "step": 375
58
  },
59
  {
60
+ "epoch": 2.13,
 
61
  "learning_rate": 0.0003,
62
+ "loss": 0.9947,
63
  "step": 400
64
  },
65
  {
66
+ "epoch": 2.67,
 
67
  "learning_rate": 0.0003,
68
+ "loss": 0.7909,
69
  "step": 500
70
  },
71
  {
72
+ "epoch": 3.0,
73
+ "eval_accuracy": 0.6093846153846154,
74
+ "eval_loss": 1.42075777053833,
75
+ "eval_runtime": 4.4832,
76
+ "eval_samples_per_second": 111.529,
77
+ "eval_steps_per_second": 14.053,
78
  "step": 562
79
  },
80
  {
81
+ "epoch": 3.0,
82
+ "eval_exact_match": 23.6,
83
+ "eval_f1": 36.76761904761906,
84
  "step": 562
85
  },
86
  {
87
  "epoch": 3.2,
 
88
  "learning_rate": 0.0003,
89
+ "loss": 0.7343,
90
  "step": 600
91
  },
92
  {
93
+ "epoch": 3.73,
 
94
  "learning_rate": 0.0003,
95
+ "loss": 0.6117,
96
  "step": 700
97
  },
98
  {
99
  "epoch": 4.0,
100
+ "eval_accuracy": 0.6065128205128205,
101
+ "eval_loss": 1.509896159172058,
102
+ "eval_runtime": 4.8185,
103
+ "eval_samples_per_second": 103.766,
104
+ "eval_steps_per_second": 13.075,
105
  "step": 750
106
  },
107
  {
108
  "epoch": 4.0,
109
+ "eval_exact_match": 22.6,
110
+ "eval_f1": 35.788253968253976,
111
  "step": 750
112
  },
113
  {
114
+ "epoch": 4.27,
 
115
  "learning_rate": 0.0003,
116
+ "loss": 0.5688,
117
  "step": 800
118
  },
119
  {
120
  "epoch": 4.8,
 
121
  "learning_rate": 0.0003,
122
+ "loss": 0.535,
123
  "step": 900
124
  },
125
  {
126
+ "epoch": 5.0,
127
+ "eval_accuracy": 0.6073333333333333,
128
+ "eval_loss": 1.6383951902389526,
129
+ "eval_runtime": 4.1044,
130
+ "eval_samples_per_second": 121.82,
131
+ "eval_steps_per_second": 15.349,
132
  "step": 937
133
  },
134
  {
135
+ "epoch": 5.0,
136
+ "eval_exact_match": 23.8,
137
+ "eval_f1": 37.04809523809523,
138
  "step": 937
139
  },
140
  {
141
+ "epoch": 5.33,
 
142
  "learning_rate": 0.0003,
143
+ "loss": 0.4997,
144
  "step": 1000
145
  },
146
  {
147
+ "epoch": 5.87,
 
148
  "learning_rate": 0.0003,
149
+ "loss": 0.4917,
150
  "step": 1100
151
  },
152
  {
153
  "epoch": 6.0,
154
+ "eval_accuracy": 0.6064102564102564,
155
+ "eval_loss": 1.6917529106140137,
156
+ "eval_runtime": 4.8299,
157
+ "eval_samples_per_second": 103.522,
158
+ "eval_steps_per_second": 13.044,
159
  "step": 1125
160
  },
161
  {
162
  "epoch": 6.0,
163
+ "eval_exact_match": 24.2,
164
+ "eval_f1": 36.030317460317455,
165
  "step": 1125
166
  },
167
  {
168
  "epoch": 6.4,
 
169
  "learning_rate": 0.0003,
170
+ "loss": 0.4605,
171
  "step": 1200
172
  },
173
  {
174
+ "epoch": 6.93,
 
175
  "learning_rate": 0.0003,
176
+ "loss": 0.4746,
177
  "step": 1300
178
  },
179
  {
180
+ "epoch": 7.0,
181
+ "eval_accuracy": 0.6048205128205129,
182
+ "eval_loss": 1.7547425031661987,
183
+ "eval_runtime": 4.2021,
184
+ "eval_samples_per_second": 118.988,
185
+ "eval_steps_per_second": 14.993,
186
  "step": 1312
187
  },
188
  {
189
+ "epoch": 7.0,
190
+ "eval_exact_match": 23.8,
191
+ "eval_f1": 36.49507936507937,
192
  "step": 1312
193
  },
194
  {
195
+ "epoch": 7.47,
 
196
  "learning_rate": 0.0003,
197
+ "loss": 0.4432,
198
  "step": 1400
199
  },
200
  {
201
  "epoch": 8.0,
 
202
  "learning_rate": 0.0003,
203
+ "loss": 0.463,
204
  "step": 1500
205
  },
206
  {
207
  "epoch": 8.0,
208
+ "eval_accuracy": 0.6056923076923076,
209
+ "eval_loss": 1.7707523107528687,
210
+ "eval_runtime": 4.821,
211
+ "eval_samples_per_second": 103.713,
212
+ "eval_steps_per_second": 13.068,
213
  "step": 1500
214
  },
215
  {
216
  "epoch": 8.0,
217
+ "eval_exact_match": 22.4,
218
+ "eval_f1": 35.834603174603174,
219
  "step": 1500
220
  },
221
  {
222
+ "epoch": 8.53,
 
223
  "learning_rate": 0.0003,
224
+ "loss": 0.4307,
225
  "step": 1600
226
  },
227
  {
228
+ "epoch": 9.0,
229
+ "eval_accuracy": 0.6051282051282051,
230
+ "eval_loss": 1.7772654294967651,
231
+ "eval_runtime": 4.4303,
232
+ "eval_samples_per_second": 112.86,
233
+ "eval_steps_per_second": 14.22,
234
  "step": 1687
235
  },
236
  {
237
+ "epoch": 9.0,
238
+ "eval_exact_match": 24.4,
239
+ "eval_f1": 36.40761904761904,
240
  "step": 1687
241
  },
242
  {
243
+ "epoch": 9.07,
 
244
  "learning_rate": 0.0003,
245
+ "loss": 0.4694,
246
  "step": 1700
247
  },
248
  {
249
  "epoch": 9.6,
 
250
  "learning_rate": 0.0003,
251
+ "loss": 0.4323,
252
  "step": 1800
253
  },
254
  {
255
  "epoch": 10.0,
256
+ "eval_accuracy": 0.5996923076923077,
257
+ "eval_loss": 1.7927325963974,
258
+ "eval_runtime": 4.4251,
259
+ "eval_samples_per_second": 112.993,
260
+ "eval_steps_per_second": 14.237,
261
  "step": 1875
262
  },
263
  {
264
  "epoch": 10.0,
265
+ "eval_exact_match": 23.2,
266
+ "eval_f1": 35.76095238095239,
267
  "step": 1875
268
  },
269
  {
270
+ "epoch": 10.13,
 
271
  "learning_rate": 0.0003,
272
+ "loss": 0.4392,
273
  "step": 1900
274
  },
275
  {
276
+ "epoch": 10.67,
 
277
  "learning_rate": 0.0003,
278
+ "loss": 0.4301,
279
  "step": 2000
280
  },
281
  {
282
+ "epoch": 11.0,
283
+ "eval_accuracy": 0.6046153846153847,
284
+ "eval_loss": 1.820131778717041,
285
+ "eval_runtime": 4.2028,
286
+ "eval_samples_per_second": 118.968,
287
+ "eval_steps_per_second": 14.99,
288
  "step": 2062
289
  },
290
  {
291
+ "epoch": 11.0,
292
+ "eval_exact_match": 22.6,
293
+ "eval_f1": 36.275425685425695,
294
  "step": 2062
295
  },
296
  {
297
  "epoch": 11.2,
 
298
  "learning_rate": 0.0003,
299
+ "loss": 0.4325,
300
  "step": 2100
301
  },
302
  {
303
+ "epoch": 11.73,
 
304
  "learning_rate": 0.0003,
305
+ "loss": 0.4331,
306
  "step": 2200
307
  },
308
  {
309
  "epoch": 12.0,
310
+ "eval_accuracy": 0.6050769230769231,
311
+ "eval_loss": 1.8902928829193115,
312
+ "eval_runtime": 4.1616,
313
+ "eval_samples_per_second": 120.145,
314
+ "eval_steps_per_second": 15.138,
315
  "step": 2250
316
  },
317
  {
318
  "epoch": 12.0,
319
+ "eval_exact_match": 23.6,
320
+ "eval_f1": 35.66952380952382,
321
  "step": 2250
322
  },
323
  {
324
+ "epoch": 12.27,
 
325
  "learning_rate": 0.0003,
326
+ "loss": 0.43,
327
  "step": 2300
328
  },
329
  {
330
  "epoch": 12.8,
 
331
  "learning_rate": 0.0003,
332
+ "loss": 0.4338,
333
  "step": 2400
334
  },
335
  {
336
+ "epoch": 13.0,
337
+ "eval_accuracy": 0.6035384615384616,
338
+ "eval_loss": 1.8489463329315186,
339
+ "eval_runtime": 4.4116,
340
+ "eval_samples_per_second": 113.338,
341
+ "eval_steps_per_second": 14.281,
342
  "step": 2437
343
  },
344
  {
345
+ "epoch": 13.0,
346
+ "eval_exact_match": 19.6,
347
+ "eval_f1": 33.558888888888895,
348
  "step": 2437
349
  }
350
  ],
351
  "logging_steps": 100,
352
  "max_steps": 9350,
 
353
  "num_train_epochs": 50,
354
  "save_steps": 500,
355
+ "total_flos": 9.094174110803558e+16,
 
 
 
 
 
 
 
 
 
 
 
 
 
356
  "trial_name": null,
357
  "trial_params": null
358
  }
checkpoint-2437/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a48aaa718d03f9c3a1cf4e80dbff36330a05dd346d42a3b722f1ae2626aabe3c
3
- size 5304
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:79a83a5d44cad230e338c4f4f6393fbd052223dad36807daa3baa278ceb8a412
3
+ size 4728