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
f9ef3ab
·
verified ·
1 Parent(s): 0e0d09a

Training in progress, epoch 20, checkpoint

Browse files
Files changed (17) hide show
  1. checkpoint-3937/README.md +9 -0
  2. checkpoint-3937/adapter_config.json +25 -0
  3. checkpoint-3937/adapter_model.safetensors +3 -0
  4. checkpoint-3937/added_tokens.json +5 -0
  5. checkpoint-3937/merges.txt +0 -0
  6. checkpoint-3937/optimizer.pt +3 -0
  7. checkpoint-3937/rng_state_0.pth +3 -0
  8. checkpoint-3937/rng_state_1.pth +3 -0
  9. checkpoint-3937/rng_state_2.pth +3 -0
  10. checkpoint-3937/rng_state_3.pth +3 -0
  11. checkpoint-3937/scheduler.pt +3 -0
  12. checkpoint-3937/special_tokens_map.json +14 -0
  13. checkpoint-3937/tokenizer.json +0 -0
  14. checkpoint-3937/tokenizer_config.json +43 -0
  15. checkpoint-3937/trainer_state.json +621 -0
  16. checkpoint-3937/training_args.bin +3 -0
  17. checkpoint-3937/vocab.json +0 -0
checkpoint-3937/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-3937/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-3937/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c5a8e0e0709f259cf6c627959af88feaaeff5da9d7b87b252732f391523e5257
3
+ size 112130416
checkpoint-3937/added_tokens.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "<|endoftext|>": 151643,
3
+ "<|im_end|>": 151645,
4
+ "<|im_start|>": 151644
5
+ }
checkpoint-3937/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-3937/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f934c33138b51072b88f643fec9e4b6089e65161994144488003313688a5d2b
3
+ size 224537202
checkpoint-3937/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:daf3a7c2de2b555ea7a4841129c9904ab61b4fa7593a657765dc5dcb26868c7c
3
+ size 15024
checkpoint-3937/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8b33741a05e5f2659805fe24777e6840294f854d62efbd4eabd383b6fc06c435
3
+ size 15024
checkpoint-3937/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ade306eba14f1ef4d258ce03cfa51a988d93e4b91363e33571fa8abcf7e666a2
3
+ size 15024
checkpoint-3937/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6b7fc74d284abda512833861374b2514c6a3fe99eb093ed9528926407682baf1
3
+ size 15024
checkpoint-3937/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1af4d103dd1539adb1a0a203dfa22ecb9652052f061b2c97e9989ac0461f70bf
3
+ size 1064
checkpoint-3937/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-3937/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-3937/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-3937/trainer_state.json ADDED
@@ -0,0 +1,621 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 20.997333333333334,
5
+ "eval_steps": 500,
6
+ "global_step": 3937,
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": 0.7527768015861511,
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
+ "epoch": 17.066666666666666,
485
+ "grad_norm": 0.7299863696098328,
486
+ "learning_rate": 0.0005,
487
+ "loss": 0.5267,
488
+ "step": 3200
489
+ },
490
+ {
491
+ "epoch": 17.6,
492
+ "grad_norm": 0.49411433935165405,
493
+ "learning_rate": 0.0005,
494
+ "loss": 0.501,
495
+ "step": 3300
496
+ },
497
+ {
498
+ "epoch": 18.0,
499
+ "eval_accuracy": 0.5468205128205128,
500
+ "eval_loss": 2.4240849018096924,
501
+ "eval_runtime": 5.0346,
502
+ "eval_samples_per_second": 99.313,
503
+ "eval_steps_per_second": 12.513,
504
+ "step": 3375
505
+ },
506
+ {
507
+ "epoch": 18.0,
508
+ "eval_exact_match": 12.8,
509
+ "eval_f1": 19.46714285714286,
510
+ "step": 3375
511
+ },
512
+ {
513
+ "epoch": 18.133333333333333,
514
+ "grad_norm": 0.5169008374214172,
515
+ "learning_rate": 0.0005,
516
+ "loss": 0.5077,
517
+ "step": 3400
518
+ },
519
+ {
520
+ "epoch": 18.666666666666668,
521
+ "grad_norm": 0.4064069092273712,
522
+ "learning_rate": 0.0005,
523
+ "loss": 0.4975,
524
+ "step": 3500
525
+ },
526
+ {
527
+ "epoch": 18.997333333333334,
528
+ "eval_accuracy": 0.5450769230769231,
529
+ "eval_loss": 2.4591097831726074,
530
+ "eval_runtime": 5.7434,
531
+ "eval_samples_per_second": 87.056,
532
+ "eval_steps_per_second": 10.969,
533
+ "step": 3562
534
+ },
535
+ {
536
+ "epoch": 18.997333333333334,
537
+ "eval_exact_match": 12.8,
538
+ "eval_f1": 19.07666666666667,
539
+ "step": 3562
540
+ },
541
+ {
542
+ "epoch": 19.2,
543
+ "grad_norm": 0.362119197845459,
544
+ "learning_rate": 0.0005,
545
+ "loss": 0.5028,
546
+ "step": 3600
547
+ },
548
+ {
549
+ "epoch": 19.733333333333334,
550
+ "grad_norm": 0.45963579416275024,
551
+ "learning_rate": 0.0005,
552
+ "loss": 0.4977,
553
+ "step": 3700
554
+ },
555
+ {
556
+ "epoch": 20.0,
557
+ "eval_accuracy": 0.545025641025641,
558
+ "eval_loss": 2.4378092288970947,
559
+ "eval_runtime": 5.0227,
560
+ "eval_samples_per_second": 99.548,
561
+ "eval_steps_per_second": 12.543,
562
+ "step": 3750
563
+ },
564
+ {
565
+ "epoch": 20.0,
566
+ "eval_exact_match": 12.6,
567
+ "eval_f1": 19.655873015873016,
568
+ "step": 3750
569
+ },
570
+ {
571
+ "epoch": 20.266666666666666,
572
+ "grad_norm": 0.6107614636421204,
573
+ "learning_rate": 0.0005,
574
+ "loss": 0.4964,
575
+ "step": 3800
576
+ },
577
+ {
578
+ "epoch": 20.8,
579
+ "grad_norm": 0.5394221544265747,
580
+ "learning_rate": 0.0005,
581
+ "loss": 0.501,
582
+ "step": 3900
583
+ },
584
+ {
585
+ "epoch": 20.997333333333334,
586
+ "eval_accuracy": 0.5438461538461539,
587
+ "eval_loss": 2.448828935623169,
588
+ "eval_runtime": 5.1176,
589
+ "eval_samples_per_second": 97.702,
590
+ "eval_steps_per_second": 12.31,
591
+ "step": 3937
592
+ },
593
+ {
594
+ "epoch": 20.997333333333334,
595
+ "eval_exact_match": 12.6,
596
+ "eval_f1": 18.367460317460317,
597
+ "step": 3937
598
+ }
599
+ ],
600
+ "logging_steps": 100,
601
+ "max_steps": 9350,
602
+ "num_input_tokens_seen": 0,
603
+ "num_train_epochs": 50,
604
+ "save_steps": 500,
605
+ "stateful_callbacks": {
606
+ "TrainerControl": {
607
+ "args": {
608
+ "should_epoch_stop": false,
609
+ "should_evaluate": false,
610
+ "should_log": false,
611
+ "should_save": true,
612
+ "should_training_stop": false
613
+ },
614
+ "attributes": {}
615
+ }
616
+ },
617
+ "total_flos": 6.857470583714611e+16,
618
+ "train_batch_size": 1,
619
+ "trial_name": null,
620
+ "trial_params": null
621
+ }
checkpoint-3937/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2776615b5656a131d464a15a3d5283d9bdf82687bc8ffc8d243c778d1158123d
3
+ size 5304
checkpoint-3937/vocab.json ADDED
The diff for this file is too large to render. See raw diff