Hugging Face's logo

ZihaoZhu
/
BoT-DeepSeek-R1-Distill-Qwen-32B

Model card Files
xet
 Community
ZihaoZhu commited on
Commit
24f08ec
·
verified ·
1 Parent(s): e5235a3

Upload trainer_state.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. trainer_state.json +313 -0
trainer_state.json ADDED
@@ -0,0 +1,313 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.9424657534246577,
5
+ "eval_steps": 500,
6
+ "global_step": 135,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.021917808219178082,
13
+ "grad_norm": 0.05345574766397476,
14
+ "learning_rate": 1.4285714285714285e-05,
15
+ "loss": 0.14798447489738464,
16
+ "memory(GiB)": 73.96,
17
+ "step": 1,
18
+ "token_acc": 0.9516081000595593,
19
+ "train_speed(iter/s)": 0.002598
20
+ },
21
+ {
22
+ "epoch": 0.1095890410958904,
23
+ "grad_norm": 0.08354049175977707,
24
+ "learning_rate": 7.142857142857143e-05,
25
+ "loss": 0.19844208657741547,
26
+ "memory(GiB)": 79.75,
27
+ "step": 5,
28
+ "token_acc": 0.9440162459039092,
29
+ "train_speed(iter/s)": 0.011652
30
+ },
31
+ {
32
+ "epoch": 0.2191780821917808,
33
+ "grad_norm": 0.09754039347171783,
34
+ "learning_rate": 9.986452283393452e-05,
35
+ "loss": 0.18643227815628052,
36
+ "memory(GiB)": 105.95,
37
+ "step": 10,
38
+ "token_acc": 0.9464405266681545,
39
+ "train_speed(iter/s)": 0.020266
40
+ },
41
+ {
42
+ "epoch": 0.3287671232876712,
43
+ "grad_norm": 0.10102561861276627,
44
+ "learning_rate": 9.903926402016153e-05,
45
+ "loss": 0.1505092740058899,
46
+ "memory(GiB)": 105.96,
47
+ "step": 15,
48
+ "token_acc": 0.9521562734785876,
49
+ "train_speed(iter/s)": 0.027262
50
+ },
51
+ {
52
+ "epoch": 0.4383561643835616,
53
+ "grad_norm": 0.14934773743152618,
54
+ "learning_rate": 9.747640902965184e-05,
55
+ "loss": 0.1314406156539917,
56
+ "memory(GiB)": 105.96,
57
+ "step": 20,
58
+ "token_acc": 0.9554180348572872,
59
+ "train_speed(iter/s)": 0.032573
60
+ },
61
+ {
62
+ "epoch": 0.547945205479452,
63
+ "grad_norm": 0.1648729294538498,
64
+ "learning_rate": 9.519946465617218e-05,
65
+ "loss": 0.1481643557548523,
66
+ "memory(GiB)": 105.96,
67
+ "step": 25,
68
+ "token_acc": 0.9479316731837827,
69
+ "train_speed(iter/s)": 0.036288
70
+ },
71
+ {
72
+ "epoch": 0.6575342465753424,
73
+ "grad_norm": 0.10335330665111542,
74
+ "learning_rate": 9.224267826248536e-05,
75
+ "loss": 0.12146763801574707,
76
+ "memory(GiB)": 105.96,
77
+ "step": 30,
78
+ "token_acc": 0.9561634805537245,
79
+ "train_speed(iter/s)": 0.040101
80
+ },
81
+ {
82
+ "epoch": 0.7671232876712328,
83
+ "grad_norm": 0.09456586837768555,
84
+ "learning_rate": 8.865052266813685e-05,
85
+ "loss": 0.1343398928642273,
86
+ "memory(GiB)": 105.96,
87
+ "step": 35,
88
+ "token_acc": 0.9518978243441241,
89
+ "train_speed(iter/s)": 0.043284
90
+ },
91
+ {
92
+ "epoch": 0.8767123287671232,
93
+ "grad_norm": 0.07303319126367569,
94
+ "learning_rate": 8.447702723685335e-05,
95
+ "loss": 0.10934320688247681,
96
+ "memory(GiB)": 115.12,
97
+ "step": 40,
98
+ "token_acc": 0.9607801568793725,
99
+ "train_speed(iter/s)": 0.0449
100
+ },
101
+ {
102
+ "epoch": 0.9863013698630136,
103
+ "grad_norm": 0.11877991259098053,
104
+ "learning_rate": 7.978496522462167e-05,
105
+ "loss": 0.126079523563385,
106
+ "memory(GiB)": 115.12,
107
+ "step": 45,
108
+ "token_acc": 0.9533266093972803,
109
+ "train_speed(iter/s)": 0.047779
110
+ },
111
+ {
112
+ "epoch": 1.0876712328767124,
113
+ "grad_norm": 0.09073251485824585,
114
+ "learning_rate": 7.464490961148921e-05,
115
+ "loss": 0.10618455410003662,
116
+ "memory(GiB)": 115.12,
117
+ "step": 50,
118
+ "token_acc": 0.9644618781415888,
119
+ "train_speed(iter/s)": 0.049565
120
+ },
121
+ {
122
+ "epoch": 1.1972602739726028,
123
+ "grad_norm": 0.08698726445436478,
124
+ "learning_rate": 6.91341716182545e-05,
125
+ "loss": 0.10942069292068482,
126
+ "memory(GiB)": 115.12,
127
+ "step": 55,
128
+ "token_acc": 0.9613163469630576,
129
+ "train_speed(iter/s)": 0.051581
130
+ },
131
+ {
132
+ "epoch": 1.3068493150684932,
133
+ "grad_norm": 0.11653894186019897,
134
+ "learning_rate": 6.333563787374493e-05,
135
+ "loss": 0.10550706386566162,
136
+ "memory(GiB)": 115.12,
137
+ "step": 60,
138
+ "token_acc": 0.964783186494069,
139
+ "train_speed(iter/s)": 0.053018
140
+ },
141
+ {
142
+ "epoch": 1.4164383561643836,
143
+ "grad_norm": 0.11223548650741577,
144
+ "learning_rate": 5.733652372276809e-05,
145
+ "loss": 0.10657280683517456,
146
+ "memory(GiB)": 115.12,
147
+ "step": 65,
148
+ "token_acc": 0.9625531401694455,
149
+ "train_speed(iter/s)": 0.054731
150
+ },
151
+ {
152
+ "epoch": 1.526027397260274,
153
+ "grad_norm": 0.14359991252422333,
154
+ "learning_rate": 5.122706142614562e-05,
155
+ "loss": 0.11491676568984985,
156
+ "memory(GiB)": 115.13,
157
+ "step": 70,
158
+ "token_acc": 0.9618089745393164,
159
+ "train_speed(iter/s)": 0.056347
160
+ },
161
+ {
162
+ "epoch": 1.6356164383561644,
163
+ "grad_norm": 0.10080211609601974,
164
+ "learning_rate": 4.509914298352197e-05,
165
+ "loss": 0.09663158655166626,
166
+ "memory(GiB)": 115.13,
167
+ "step": 75,
168
+ "token_acc": 0.965161357921207,
169
+ "train_speed(iter/s)": 0.057444
170
+ },
171
+ {
172
+ "epoch": 1.7452054794520548,
173
+ "grad_norm": 0.0964246392250061,
174
+ "learning_rate": 3.9044937992156516e-05,
175
+ "loss": 0.10549492835998535,
176
+ "memory(GiB)": 115.13,
177
+ "step": 80,
178
+ "token_acc": 0.9612994018998475,
179
+ "train_speed(iter/s)": 0.058281
180
+ },
181
+ {
182
+ "epoch": 1.8547945205479452,
183
+ "grad_norm": 0.09354526549577713,
184
+ "learning_rate": 3.3155507330389e-05,
185
+ "loss": 0.08785929679870605,
186
+ "memory(GiB)": 115.13,
187
+ "step": 85,
188
+ "token_acc": 0.9670452769238785,
189
+ "train_speed(iter/s)": 0.059315
190
+ },
191
+ {
192
+ "epoch": 1.9643835616438357,
193
+ "grad_norm": 0.12421014159917831,
194
+ "learning_rate": 2.7519433517269665e-05,
195
+ "loss": 0.10144306421279907,
196
+ "memory(GiB)": 115.13,
197
+ "step": 90,
198
+ "token_acc": 0.9624555254633317,
199
+ "train_speed(iter/s)": 0.060163
200
+ },
201
+ {
202
+ "epoch": 2.0657534246575344,
203
+ "grad_norm": 0.08086857199668884,
204
+ "learning_rate": 2.2221488349019903e-05,
205
+ "loss": 0.08750819563865661,
206
+ "memory(GiB)": 115.13,
207
+ "step": 95,
208
+ "token_acc": 0.9699472974629244,
209
+ "train_speed(iter/s)": 0.06089
210
+ },
211
+ {
212
+ "epoch": 2.175342465753425,
213
+ "grad_norm": 0.12894976139068604,
214
+ "learning_rate": 1.7341357852311174e-05,
215
+ "loss": 0.09126536846160889,
216
+ "memory(GiB)": 115.13,
217
+ "step": 100,
218
+ "token_acc": 0.9701614366043896,
219
+ "train_speed(iter/s)": 0.061361
220
+ },
221
+ {
222
+ "epoch": 2.2849315068493152,
223
+ "grad_norm": 0.14563050866127014,
224
+ "learning_rate": 1.2952443732252057e-05,
225
+ "loss": 0.08330788612365722,
226
+ "memory(GiB)": 115.13,
227
+ "step": 105,
228
+ "token_acc": 0.9722385948367323,
229
+ "train_speed(iter/s)": 0.062287
230
+ },
231
+ {
232
+ "epoch": 2.3945205479452056,
233
+ "grad_norm": 0.1167457178235054,
234
+ "learning_rate": 9.12075934242082e-06,
235
+ "loss": 0.08607174158096313,
236
+ "memory(GiB)": 115.13,
237
+ "step": 110,
238
+ "token_acc": 0.9686517477344183,
239
+ "train_speed(iter/s)": 0.063146
240
+ },
241
+ {
242
+ "epoch": 2.504109589041096,
243
+ "grad_norm": 0.10588780790567398,
244
+ "learning_rate": 5.903936782582253e-06,
245
+ "loss": 0.07531117200851441,
246
+ "memory(GiB)": 115.13,
247
+ "step": 115,
248
+ "token_acc": 0.9729850523911209,
249
+ "train_speed(iter/s)": 0.06371
250
+ },
251
+ {
252
+ "epoch": 2.6136986301369864,
253
+ "grad_norm": 0.1297299861907959,
254
+ "learning_rate": 3.350360058263058e-06,
255
+ "loss": 0.08095236420631409,
256
+ "memory(GiB)": 115.13,
257
+ "step": 120,
258
+ "token_acc": 0.9723439137796651,
259
+ "train_speed(iter/s)": 0.064274
260
+ },
261
+ {
262
+ "epoch": 2.723287671232877,
263
+ "grad_norm": 0.11432217061519623,
264
+ "learning_rate": 1.4984373402728014e-06,
265
+ "loss": 0.09324527382850648,
266
+ "memory(GiB)": 115.15,
267
+ "step": 125,
268
+ "token_acc": 0.9666533115542268,
269
+ "train_speed(iter/s)": 0.064962
270
+ },
271
+ {
272
+ "epoch": 2.8328767123287673,
273
+ "grad_norm": 0.10140057653188705,
274
+ "learning_rate": 3.7602327006450167e-07,
275
+ "loss": 0.07901231050491334,
276
+ "memory(GiB)": 115.15,
277
+ "step": 130,
278
+ "token_acc": 0.9727071738964702,
279
+ "train_speed(iter/s)": 0.065426
280
+ },
281
+ {
282
+ "epoch": 2.9424657534246577,
283
+ "grad_norm": 0.09595915675163269,
284
+ "learning_rate": 0.0,
285
+ "loss": 0.08781038522720337,
286
+ "memory(GiB)": 115.15,
287
+ "step": 135,
288
+ "token_acc": 0.9682683590208522,
289
+ "train_speed(iter/s)": 0.065788
290
+ }
291
+ ],
292
+ "logging_steps": 5,
293
+ "max_steps": 135,
294
+ "num_input_tokens_seen": 0,
295
+ "num_train_epochs": 3,
296
+ "save_steps": 500,
297
+ "stateful_callbacks": {
298
+ "TrainerControl": {
299
+ "args": {
300
+ "should_epoch_stop": false,
301
+ "should_evaluate": false,
302
+ "should_log": false,
303
+ "should_save": true,
304
+ "should_training_stop": true
305
+ },
306
+ "attributes": {}
307
+ }
308
+ },
309
+ "total_flos": 2.1292311990945178e+17,
310
+ "train_batch_size": 1,
311
+ "trial_name": null,
312
+ "trial_params": null
313
+ }