Hugging Face's logo

tyzhu
/
lmind_nq_train6000_eval6489_v1_qa_1e-4_lora2

Safetensors
Generated from Trainer
Eval Results
Model card Files Community
tyzhu commited on
Commit
b98c33d
•
1 Parent(s): 77c18e2

Training in progress, epoch 16, checkpoint

Browse files
Files changed (17) hide show
  1. checkpoint-3000/README.md +9 -0
  2. checkpoint-3000/adapter_config.json +25 -0
  3. checkpoint-3000/adapter_model.safetensors +3 -0
  4. checkpoint-3000/added_tokens.json +5 -0
  5. checkpoint-3000/merges.txt +0 -0
  6. checkpoint-3000/optimizer.pt +3 -0
  7. checkpoint-3000/rng_state_0.pth +3 -0
  8. checkpoint-3000/rng_state_1.pth +3 -0
  9. checkpoint-3000/rng_state_2.pth +3 -0
  10. checkpoint-3000/rng_state_3.pth +3 -0
  11. checkpoint-3000/scheduler.pt +3 -0
  12. checkpoint-3000/special_tokens_map.json +14 -0
  13. checkpoint-3000/tokenizer.json +0 -0
  14. checkpoint-3000/tokenizer_config.json +43 -0
  15. checkpoint-3000/trainer_state.json +483 -0
  16. checkpoint-3000/training_args.bin +3 -0
  17. checkpoint-3000/vocab.json +0 -0
checkpoint-3000/README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+
9
+ - PEFT 0.5.0
checkpoint-3000/adapter_config.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 16,
11
+ "lora_dropout": 0.05,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 16,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "k_proj",
19
+ "v_proj",
20
+ "down_proj",
21
+ "gate_proj",
22
+ "up_proj"
23
+ ],
24
+ "task_type": "CAUSAL_LM"
25
+ }
checkpoint-3000/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b12b97096035324db5235591ac828c78ab9021ce3c8555e0defacd3419fa50a5
3
+ size 112130416
checkpoint-3000/added_tokens.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "<|endoftext|>": 151643,
3
+ "<|im_end|>": 151645,
4
+ "<|im_start|>": 151644
5
+ }
checkpoint-3000/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-3000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:051b2a87191be3450033167e4a1810e29811916257297c1764c6ca18cf714692
3
+ size 224537202
checkpoint-3000/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6dfffb25ae46fc066e4e1dabc512cb0d91a52cbbe54d9f67506af4534fec54c6
3
+ size 15024
checkpoint-3000/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e8db34e447497bdb6a1743a6ec01332ed3271d6eeeb788331ac25de67e491ae
3
+ size 15024
checkpoint-3000/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6a19b102c820abda4d85d539fb4088fe07d2dea3c61315a442b66357a1bee48c
3
+ size 15024
checkpoint-3000/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3d4c095e2baf66535e2ace827a399f7562836960cf3dc57c502d48aeca0b660c
3
+ size 15024
checkpoint-3000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:61725e453643aca3b98d3f6fe59cc89762ac80a6e72955656f8de7c9fa16e931
3
+ size 1064
checkpoint-3000/special_tokens_map.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ }
checkpoint-3000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-3000/tokenizer_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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,
25
+ "single_word": false,
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
+ }
checkpoint-3000/trainer_state.json ADDED
@@ -0,0 +1,483 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 16.0,
5
+ "eval_steps": 500,
6
+ "global_step": 3000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.5333333333333333,
13
+ "grad_norm": 0.64143306016922,
14
+ "learning_rate": 0.0001,
15
+ "loss": 1.7779,
16
+ "step": 100
17
+ },
18
+ {
19
+ "epoch": 0.9973333333333333,
20
+ "eval_accuracy": 0.5734871794871795,
21
+ "eval_loss": 1.6191272735595703,
22
+ "eval_runtime": 5.393,
23
+ "eval_samples_per_second": 92.713,
24
+ "eval_steps_per_second": 11.682,
25
+ "step": 187
26
+ },
27
+ {
28
+ "epoch": 0.9973333333333333,
29
+ "eval_exact_match": 18.0,
30
+ "eval_f1": 28.280476190476193,
31
+ "step": 187
32
+ },
33
+ {
34
+ "epoch": 1.0666666666666667,
35
+ "grad_norm": 0.7118939161300659,
36
+ "learning_rate": 0.0001,
37
+ "loss": 1.6349,
38
+ "step": 200
39
+ },
40
+ {
41
+ "epoch": 1.6,
42
+ "grad_norm": 1.0694799423217773,
43
+ "learning_rate": 0.0001,
44
+ "loss": 1.4887,
45
+ "step": 300
46
+ },
47
+ {
48
+ "epoch": 2.0,
49
+ "eval_accuracy": 0.5735897435897436,
50
+ "eval_loss": 1.6172032356262207,
51
+ "eval_runtime": 5.0425,
52
+ "eval_samples_per_second": 99.158,
53
+ "eval_steps_per_second": 12.494,
54
+ "step": 375
55
+ },
56
+ {
57
+ "epoch": 2.0,
58
+ "eval_exact_match": 19.8,
59
+ "eval_f1": 28.77095238095238,
60
+ "step": 375
61
+ },
62
+ {
63
+ "epoch": 2.1333333333333333,
64
+ "grad_norm": 1.28153395652771,
65
+ "learning_rate": 0.0001,
66
+ "loss": 1.4304,
67
+ "step": 400
68
+ },
69
+ {
70
+ "epoch": 2.6666666666666665,
71
+ "grad_norm": 1.3190444707870483,
72
+ "learning_rate": 0.0001,
73
+ "loss": 1.2346,
74
+ "step": 500
75
+ },
76
+ {
77
+ "epoch": 2.997333333333333,
78
+ "eval_accuracy": 0.5706153846153846,
79
+ "eval_loss": 1.6950472593307495,
80
+ "eval_runtime": 5.1981,
81
+ "eval_samples_per_second": 96.188,
82
+ "eval_steps_per_second": 12.12,
83
+ "step": 562
84
+ },
85
+ {
86
+ "epoch": 2.997333333333333,
87
+ "eval_exact_match": 19.4,
88
+ "eval_f1": 28.241904761904756,
89
+ "step": 562
90
+ },
91
+ {
92
+ "epoch": 3.2,
93
+ "grad_norm": 1.5040416717529297,
94
+ "learning_rate": 0.0001,
95
+ "loss": 1.1583,
96
+ "step": 600
97
+ },
98
+ {
99
+ "epoch": 3.7333333333333334,
100
+ "grad_norm": 1.512290358543396,
101
+ "learning_rate": 0.0001,
102
+ "loss": 1.0104,
103
+ "step": 700
104
+ },
105
+ {
106
+ "epoch": 4.0,
107
+ "eval_accuracy": 0.5655897435897436,
108
+ "eval_loss": 1.7944799661636353,
109
+ "eval_runtime": 5.3441,
110
+ "eval_samples_per_second": 93.561,
111
+ "eval_steps_per_second": 11.789,
112
+ "step": 750
113
+ },
114
+ {
115
+ "epoch": 4.0,
116
+ "eval_exact_match": 15.8,
117
+ "eval_f1": 23.639523809523823,
118
+ "step": 750
119
+ },
120
+ {
121
+ "epoch": 4.266666666666667,
122
+ "grad_norm": 1.698282241821289,
123
+ "learning_rate": 0.0001,
124
+ "loss": 0.9275,
125
+ "step": 800
126
+ },
127
+ {
128
+ "epoch": 4.8,
129
+ "grad_norm": 1.6007401943206787,
130
+ "learning_rate": 0.0001,
131
+ "loss": 0.843,
132
+ "step": 900
133
+ },
134
+ {
135
+ "epoch": 4.997333333333334,
136
+ "eval_accuracy": 0.5632820512820513,
137
+ "eval_loss": 1.9219905138015747,
138
+ "eval_runtime": 5.2659,
139
+ "eval_samples_per_second": 94.951,
140
+ "eval_steps_per_second": 11.964,
141
+ "step": 937
142
+ },
143
+ {
144
+ "epoch": 4.997333333333334,
145
+ "eval_exact_match": 15.8,
146
+ "eval_f1": 24.631904761904778,
147
+ "step": 937
148
+ },
149
+ {
150
+ "epoch": 5.333333333333333,
151
+ "grad_norm": 1.5375592708587646,
152
+ "learning_rate": 0.0001,
153
+ "loss": 0.7516,
154
+ "step": 1000
155
+ },
156
+ {
157
+ "epoch": 5.866666666666667,
158
+ "grad_norm": 1.718989372253418,
159
+ "learning_rate": 0.0001,
160
+ "loss": 0.7199,
161
+ "step": 1100
162
+ },
163
+ {
164
+ "epoch": 6.0,
165
+ "eval_accuracy": 0.5616923076923077,
166
+ "eval_loss": 2.016322612762451,
167
+ "eval_runtime": 5.116,
168
+ "eval_samples_per_second": 97.734,
169
+ "eval_steps_per_second": 12.314,
170
+ "step": 1125
171
+ },
172
+ {
173
+ "epoch": 6.0,
174
+ "eval_exact_match": 14.4,
175
+ "eval_f1": 22.39285714285715,
176
+ "step": 1125
177
+ },
178
+ {
179
+ "epoch": 6.4,
180
+ "grad_norm": 1.8337936401367188,
181
+ "learning_rate": 0.0001,
182
+ "loss": 0.6378,
183
+ "step": 1200
184
+ },
185
+ {
186
+ "epoch": 6.933333333333334,
187
+ "grad_norm": 1.9019217491149902,
188
+ "learning_rate": 0.0001,
189
+ "loss": 0.6432,
190
+ "step": 1300
191
+ },
192
+ {
193
+ "epoch": 6.997333333333334,
194
+ "eval_accuracy": 0.5616410256410257,
195
+ "eval_loss": 2.1283841133117676,
196
+ "eval_runtime": 5.1352,
197
+ "eval_samples_per_second": 97.367,
198
+ "eval_steps_per_second": 12.268,
199
+ "step": 1312
200
+ },
201
+ {
202
+ "epoch": 6.997333333333334,
203
+ "eval_exact_match": 13.8,
204
+ "eval_f1": 22.042857142857148,
205
+ "step": 1312
206
+ },
207
+ {
208
+ "epoch": 7.466666666666667,
209
+ "grad_norm": 1.246907114982605,
210
+ "learning_rate": 0.0001,
211
+ "loss": 0.578,
212
+ "step": 1400
213
+ },
214
+ {
215
+ "epoch": 8.0,
216
+ "grad_norm": 1.6256760358810425,
217
+ "learning_rate": 0.0001,
218
+ "loss": 0.596,
219
+ "step": 1500
220
+ },
221
+ {
222
+ "epoch": 8.0,
223
+ "eval_accuracy": 0.5603589743589743,
224
+ "eval_loss": 2.170966386795044,
225
+ "eval_runtime": 5.0636,
226
+ "eval_samples_per_second": 98.744,
227
+ "eval_steps_per_second": 12.442,
228
+ "step": 1500
229
+ },
230
+ {
231
+ "epoch": 8.0,
232
+ "eval_exact_match": 14.4,
233
+ "eval_f1": 21.71636363636364,
234
+ "step": 1500
235
+ },
236
+ {
237
+ "epoch": 8.533333333333333,
238
+ "grad_norm": 1.3106721639633179,
239
+ "learning_rate": 0.0001,
240
+ "loss": 0.543,
241
+ "step": 1600
242
+ },
243
+ {
244
+ "epoch": 8.997333333333334,
245
+ "eval_accuracy": 0.5595384615384615,
246
+ "eval_loss": 2.262348175048828,
247
+ "eval_runtime": 5.3198,
248
+ "eval_samples_per_second": 93.989,
249
+ "eval_steps_per_second": 11.843,
250
+ "step": 1687
251
+ },
252
+ {
253
+ "epoch": 8.997333333333334,
254
+ "eval_exact_match": 16.0,
255
+ "eval_f1": 24.34507936507938,
256
+ "step": 1687
257
+ },
258
+ {
259
+ "epoch": 9.066666666666666,
260
+ "grad_norm": 1.0499460697174072,
261
+ "learning_rate": 0.0001,
262
+ "loss": 0.5644,
263
+ "step": 1700
264
+ },
265
+ {
266
+ "epoch": 9.6,
267
+ "grad_norm": 0.9217801690101624,
268
+ "learning_rate": 0.0001,
269
+ "loss": 0.5314,
270
+ "step": 1800
271
+ },
272
+ {
273
+ "epoch": 10.0,
274
+ "eval_accuracy": 0.5583589743589744,
275
+ "eval_loss": 2.268125057220459,
276
+ "eval_runtime": 5.457,
277
+ "eval_samples_per_second": 91.626,
278
+ "eval_steps_per_second": 11.545,
279
+ "step": 1875
280
+ },
281
+ {
282
+ "epoch": 10.0,
283
+ "eval_exact_match": 15.4,
284
+ "eval_f1": 23.46095238095238,
285
+ "step": 1875
286
+ },
287
+ {
288
+ "epoch": 10.133333333333333,
289
+ "grad_norm": 0.969529926776886,
290
+ "learning_rate": 0.0001,
291
+ "loss": 0.534,
292
+ "step": 1900
293
+ },
294
+ {
295
+ "epoch": 10.666666666666666,
296
+ "grad_norm": 0.9540187120437622,
297
+ "learning_rate": 0.0001,
298
+ "loss": 0.5171,
299
+ "step": 2000
300
+ },
301
+ {
302
+ "epoch": 10.997333333333334,
303
+ "eval_accuracy": 0.5584615384615385,
304
+ "eval_loss": 2.3312978744506836,
305
+ "eval_runtime": 5.2187,
306
+ "eval_samples_per_second": 95.81,
307
+ "eval_steps_per_second": 12.072,
308
+ "step": 2062
309
+ },
310
+ {
311
+ "epoch": 10.997333333333334,
312
+ "eval_exact_match": 16.2,
313
+ "eval_f1": 23.968888888888895,
314
+ "step": 2062
315
+ },
316
+ {
317
+ "epoch": 11.2,
318
+ "grad_norm": 0.8946124911308289,
319
+ "learning_rate": 0.0001,
320
+ "loss": 0.5208,
321
+ "step": 2100
322
+ },
323
+ {
324
+ "epoch": 11.733333333333333,
325
+ "grad_norm": 0.8362706899642944,
326
+ "learning_rate": 0.0001,
327
+ "loss": 0.512,
328
+ "step": 2200
329
+ },
330
+ {
331
+ "epoch": 12.0,
332
+ "eval_accuracy": 0.559025641025641,
333
+ "eval_loss": 2.364483594894409,
334
+ "eval_runtime": 5.0957,
335
+ "eval_samples_per_second": 98.122,
336
+ "eval_steps_per_second": 12.363,
337
+ "step": 2250
338
+ },
339
+ {
340
+ "epoch": 12.0,
341
+ "eval_exact_match": 16.0,
342
+ "eval_f1": 24.015714285714285,
343
+ "step": 2250
344
+ },
345
+ {
346
+ "epoch": 12.266666666666667,
347
+ "grad_norm": 0.7387551665306091,
348
+ "learning_rate": 0.0001,
349
+ "loss": 0.5038,
350
+ "step": 2300
351
+ },
352
+ {
353
+ "epoch": 12.8,
354
+ "grad_norm": 0.8488346338272095,
355
+ "learning_rate": 0.0001,
356
+ "loss": 0.503,
357
+ "step": 2400
358
+ },
359
+ {
360
+ "epoch": 12.997333333333334,
361
+ "eval_accuracy": 0.5588205128205128,
362
+ "eval_loss": 2.3642191886901855,
363
+ "eval_runtime": 5.4807,
364
+ "eval_samples_per_second": 91.23,
365
+ "eval_steps_per_second": 11.495,
366
+ "step": 2437
367
+ },
368
+ {
369
+ "epoch": 12.997333333333334,
370
+ "eval_exact_match": 16.0,
371
+ "eval_f1": 23.92238095238096,
372
+ "step": 2437
373
+ },
374
+ {
375
+ "epoch": 13.333333333333334,
376
+ "grad_norm": 0.9431668519973755,
377
+ "learning_rate": 0.0001,
378
+ "loss": 0.4941,
379
+ "step": 2500
380
+ },
381
+ {
382
+ "epoch": 13.866666666666667,
383
+ "grad_norm": 0.8775368928909302,
384
+ "learning_rate": 0.0001,
385
+ "loss": 0.4996,
386
+ "step": 2600
387
+ },
388
+ {
389
+ "epoch": 14.0,
390
+ "eval_accuracy": 0.5582051282051282,
391
+ "eval_loss": 2.436706304550171,
392
+ "eval_runtime": 5.1472,
393
+ "eval_samples_per_second": 97.141,
394
+ "eval_steps_per_second": 12.24,
395
+ "step": 2625
396
+ },
397
+ {
398
+ "epoch": 14.0,
399
+ "eval_exact_match": 15.4,
400
+ "eval_f1": 24.02761904761905,
401
+ "step": 2625
402
+ },
403
+ {
404
+ "epoch": 14.4,
405
+ "grad_norm": 0.6873505115509033,
406
+ "learning_rate": 0.0001,
407
+ "loss": 0.4835,
408
+ "step": 2700
409
+ },
410
+ {
411
+ "epoch": 14.933333333333334,
412
+ "grad_norm": 0.7216406464576721,
413
+ "learning_rate": 0.0001,
414
+ "loss": 0.4966,
415
+ "step": 2800
416
+ },
417
+ {
418
+ "epoch": 14.997333333333334,
419
+ "eval_accuracy": 0.5585128205128205,
420
+ "eval_loss": 2.4315266609191895,
421
+ "eval_runtime": 5.7766,
422
+ "eval_samples_per_second": 86.556,
423
+ "eval_steps_per_second": 10.906,
424
+ "step": 2812
425
+ },
426
+ {
427
+ "epoch": 14.997333333333334,
428
+ "eval_exact_match": 15.2,
429
+ "eval_f1": 23.853333333333342,
430
+ "step": 2812
431
+ },
432
+ {
433
+ "epoch": 15.466666666666667,
434
+ "grad_norm": 0.7105251550674438,
435
+ "learning_rate": 0.0001,
436
+ "loss": 0.476,
437
+ "step": 2900
438
+ },
439
+ {
440
+ "epoch": 16.0,
441
+ "grad_norm": 0.6648848056793213,
442
+ "learning_rate": 0.0001,
443
+ "loss": 0.4967,
444
+ "step": 3000
445
+ },
446
+ {
447
+ "epoch": 16.0,
448
+ "eval_accuracy": 0.5587179487179487,
449
+ "eval_loss": 2.417990207672119,
450
+ "eval_runtime": 5.286,
451
+ "eval_samples_per_second": 94.589,
452
+ "eval_steps_per_second": 11.918,
453
+ "step": 3000
454
+ },
455
+ {
456
+ "epoch": 16.0,
457
+ "eval_exact_match": 14.8,
458
+ "eval_f1": 23.16634920634922,
459
+ "step": 3000
460
+ }
461
+ ],
462
+ "logging_steps": 100,
463
+ "max_steps": 9350,
464
+ "num_input_tokens_seen": 0,
465
+ "num_train_epochs": 50,
466
+ "save_steps": 500,
467
+ "stateful_callbacks": {
468
+ "TrainerControl": {
469
+ "args": {
470
+ "should_epoch_stop": false,
471
+ "should_evaluate": false,
472
+ "should_log": false,
473
+ "should_save": true,
474
+ "should_training_stop": false
475
+ },
476
+ "attributes": {}
477
+ }
478
+ },
479
+ "total_flos": 5.495238963914342e+16,
480
+ "train_batch_size": 2,
481
+ "trial_name": null,
482
+ "trial_params": null
483
+ }
checkpoint-3000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d667d5e0e40fe951f9fe0d2cd7fb4682a608b09af78a113a245307211adedfbc
3
+ size 5304
checkpoint-3000/vocab.json ADDED
The diff for this file is too large to render. See raw diff