Blancy commited on
Commit
f0a1ba0
·
verified ·
1 Parent(s): 08970f8

Training in progress, step 100

Browse files
Files changed (7) hide show
  1. README.md +4 -4
  2. all_results.json +6 -6
  3. model.safetensors +1 -1
  4. task.sh +5 -0
  5. train_results.json +6 -6
  6. trainer_state.json +294 -126
  7. training_args.bin +2 -2
README.md CHANGED
@@ -27,18 +27,18 @@ print(output["generated_text"])
27
 
28
  ## Training procedure
29
 
30
- [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/224015062-chinese-university-of-hong-kong-shenzhen/huggingface/runs/j2a2ufcx)
31
 
32
 
33
  This model was trained with SFT.
34
 
35
  ### Framework versions
36
 
37
- - TRL: 0.15.2
38
  - Transformers: 4.49.0
39
  - Pytorch: 2.5.1
40
- - Datasets: 3.3.2
41
- - Tokenizers: 0.21.0
42
 
43
  ## Citations
44
 
 
27
 
28
  ## Training procedure
29
 
30
+ [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/224015062-chinese-university-of-hong-kong-shenzhen/huggingface/runs/7ehmga5f)
31
 
32
 
33
  This model was trained with SFT.
34
 
35
  ### Framework versions
36
 
37
+ - TRL: 0.16.0.dev0
38
  - Transformers: 4.49.0
39
  - Pytorch: 2.5.1
40
+ - Datasets: 3.5.0
41
+ - Tokenizers: 0.21.1
42
 
43
  ## Citations
44
 
all_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "total_flos": 8261436702720.0,
3
- "train_loss": 1.3481066539369781,
4
- "train_runtime": 309.1787,
5
- "train_samples": 1000,
6
- "train_samples_per_second": 29.921,
7
- "train_steps_per_second": 0.469
8
  }
 
1
  {
2
+ "total_flos": 4.4183709150420664e+18,
3
+ "train_loss": 0.07813258930595952,
4
+ "train_runtime": 6469.3467,
5
+ "train_samples": 93733,
6
+ "train_samples_per_second": 5.302,
7
+ "train_steps_per_second": 0.041
8
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0ba5e14bf244f141663a159f7effd49452abbff4ab13dcf021aed5b9bccd11c5
3
  size 3087467144
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d16895f9a5175083b38e5654923cd89c0a5ebbcfd659f54779d49fc8cc2a55a
3
  size 3087467144
task.sh ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ cd /dengwenbo/224015062/open-r1-main
2
+ wandb login fcfbe03316742987758317ffb3e13a1eda483ecc
3
+ ACCELERATE_LOG_LEVEL=info
4
+ accelerate launch --config_file recipes/accelerate_configs/zero3.yaml
5
+ python src/open_r1/sft.py --config recipes/Qwen2.5-1.5B-Instruct/sft/config_demo.yaml
train_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "total_flos": 8261436702720.0,
3
- "train_loss": 1.3481066539369781,
4
- "train_runtime": 309.1787,
5
- "train_samples": 1000,
6
- "train_samples_per_second": 29.921,
7
- "train_steps_per_second": 0.469
8
  }
 
1
  {
2
+ "total_flos": 4.4183709150420664e+18,
3
+ "train_loss": 0.07813258930595952,
4
+ "train_runtime": 6469.3467,
5
+ "train_samples": 93733,
6
+ "train_samples_per_second": 5.302,
7
+ "train_steps_per_second": 0.041
8
  }
trainer_state.json CHANGED
@@ -3,226 +3,394 @@
3
  "best_model_checkpoint": null,
4
  "epoch": 1.0,
5
  "eval_steps": 500,
6
- "global_step": 145,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
- "epoch": 0.034482758620689655,
13
- "grad_norm": 2.223879312740725,
14
- "learning_rate": 3.125e-05,
15
- "loss": 1.6124,
16
  "step": 5
17
  },
18
  {
19
- "epoch": 0.06896551724137931,
20
- "grad_norm": 1.8593462844920996,
21
- "learning_rate": 4.997634107543713e-05,
22
- "loss": 1.483,
23
  "step": 10
24
  },
25
  {
26
- "epoch": 0.10344827586206896,
27
- "grad_norm": 1.2190914332382117,
28
- "learning_rate": 4.971074924974395e-05,
29
- "loss": 1.4303,
30
  "step": 15
31
  },
32
  {
33
- "epoch": 0.13793103448275862,
34
- "grad_norm": 1.0812975621686058,
35
- "learning_rate": 4.915349129238729e-05,
36
- "loss": 1.4119,
37
  "step": 20
38
  },
39
  {
40
- "epoch": 0.1724137931034483,
41
- "grad_norm": 1.0386501468781904,
42
- "learning_rate": 4.8311884978689945e-05,
43
- "loss": 1.3904,
44
  "step": 25
45
  },
46
  {
47
- "epoch": 0.20689655172413793,
48
- "grad_norm": 0.8981832087913736,
49
- "learning_rate": 4.719698207710602e-05,
50
- "loss": 1.3854,
51
  "step": 30
52
  },
53
  {
54
- "epoch": 0.2413793103448276,
55
- "grad_norm": 0.9863245505204763,
56
- "learning_rate": 4.582342322009812e-05,
57
- "loss": 1.3706,
58
  "step": 35
59
  },
60
  {
61
- "epoch": 0.27586206896551724,
62
- "grad_norm": 0.8723743249631993,
63
- "learning_rate": 4.4209245646929606e-05,
64
- "loss": 1.377,
65
  "step": 40
66
  },
67
  {
68
- "epoch": 0.3103448275862069,
69
- "grad_norm": 0.9389609365886089,
70
- "learning_rate": 4.2375646343046135e-05,
71
- "loss": 1.354,
72
  "step": 45
73
  },
74
  {
75
- "epoch": 0.3448275862068966,
76
- "grad_norm": 0.9078155340095996,
77
- "learning_rate": 4.034670368644256e-05,
78
- "loss": 1.37,
79
  "step": 50
80
  },
81
  {
82
- "epoch": 0.3793103448275862,
83
- "grad_norm": 0.8030698620747992,
84
- "learning_rate": 3.8149061256287007e-05,
85
- "loss": 1.374,
86
  "step": 55
87
  },
88
  {
89
- "epoch": 0.41379310344827586,
90
- "grad_norm": 0.8467334234236363,
91
- "learning_rate": 3.581157795594989e-05,
92
- "loss": 1.3178,
93
  "step": 60
94
  },
95
  {
96
- "epoch": 0.4482758620689655,
97
- "grad_norm": 0.8223111597770548,
98
- "learning_rate": 3.3364949044936924e-05,
99
- "loss": 1.3572,
100
  "step": 65
101
  },
102
  {
103
- "epoch": 0.4827586206896552,
104
- "grad_norm": 0.8325338686878566,
105
- "learning_rate": 3.084130305624209e-05,
106
- "loss": 1.3393,
107
  "step": 70
108
  },
109
  {
110
- "epoch": 0.5172413793103449,
111
- "grad_norm": 0.8266000883686835,
112
- "learning_rate": 2.8273779892303337e-05,
113
- "loss": 1.2754,
114
  "step": 75
115
  },
116
  {
117
- "epoch": 0.5517241379310345,
118
- "grad_norm": 0.7826202562945184,
119
- "learning_rate": 2.5696095639901996e-05,
120
- "loss": 1.3357,
121
  "step": 80
122
  },
123
  {
124
- "epoch": 0.5862068965517241,
125
- "grad_norm": 0.7673219689160695,
126
- "learning_rate": 2.314209981875008e-05,
127
- "loss": 1.349,
128
  "step": 85
129
  },
130
  {
131
- "epoch": 0.6206896551724138,
132
- "grad_norm": 0.7415260497592828,
133
- "learning_rate": 2.0645330877869e-05,
134
- "loss": 1.3149,
135
  "step": 90
136
  },
137
  {
138
- "epoch": 0.6551724137931034,
139
- "grad_norm": 0.743723421645456,
140
- "learning_rate": 1.8238575776872595e-05,
141
- "loss": 1.2909,
142
  "step": 95
143
  },
144
  {
145
- "epoch": 0.6896551724137931,
146
- "grad_norm": 0.7636915735540463,
147
- "learning_rate": 1.5953439435625384e-05,
148
- "loss": 1.311,
149
  "step": 100
150
  },
151
  {
152
- "epoch": 0.7241379310344828,
153
- "grad_norm": 0.7633473936950439,
154
- "learning_rate": 1.3819929706157959e-05,
155
- "loss": 1.3251,
156
  "step": 105
157
  },
158
  {
159
- "epoch": 0.7586206896551724,
160
- "grad_norm": 0.7474290479644277,
161
- "learning_rate": 1.1866063316886964e-05,
162
- "loss": 1.3102,
163
  "step": 110
164
  },
165
  {
166
- "epoch": 0.7931034482758621,
167
- "grad_norm": 0.7520390376697367,
168
- "learning_rate": 1.0117497963783762e-05,
169
- "loss": 1.2939,
170
  "step": 115
171
  },
172
  {
173
- "epoch": 0.8275862068965517,
174
- "grad_norm": 0.7145745995851329,
175
- "learning_rate": 8.597195379780726e-06,
176
- "loss": 1.3224,
177
  "step": 120
178
  },
179
  {
180
- "epoch": 0.8620689655172413,
181
- "grad_norm": 0.6935384011256254,
182
- "learning_rate": 7.3251198069053465e-06,
183
- "loss": 1.2729,
184
  "step": 125
185
  },
186
  {
187
- "epoch": 0.896551724137931,
188
- "grad_norm": 0.68511709825691,
189
- "learning_rate": 6.317975830732497e-06,
190
- "loss": 1.2847,
191
  "step": 130
192
  },
193
  {
194
- "epoch": 0.9310344827586207,
195
- "grad_norm": 0.6717353850255624,
196
- "learning_rate": 5.588989019848609e-06,
197
- "loss": 1.2687,
198
  "step": 135
199
  },
200
  {
201
- "epoch": 0.9655172413793104,
202
- "grad_norm": 0.7112591296225297,
203
- "learning_rate": 5.147732250916841e-06,
204
- "loss": 1.3043,
205
  "step": 140
206
  },
207
  {
208
- "epoch": 1.0,
209
- "grad_norm": 0.7123736228791405,
210
- "learning_rate": 5e-06,
211
- "loss": 1.2627,
212
  "step": 145
213
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
214
  {
215
  "epoch": 1.0,
216
- "step": 145,
217
- "total_flos": 8261436702720.0,
218
- "train_loss": 1.3481066539369781,
219
- "train_runtime": 309.1787,
220
- "train_samples_per_second": 29.921,
221
- "train_steps_per_second": 0.469
222
  }
223
  ],
224
  "logging_steps": 5,
225
- "max_steps": 145,
226
  "num_input_tokens_seen": 0,
227
  "num_train_epochs": 1,
228
  "save_steps": 100,
@@ -238,8 +406,8 @@
238
  "attributes": {}
239
  }
240
  },
241
- "total_flos": 8261436702720.0,
242
- "train_batch_size": 16,
243
  "trial_name": null,
244
  "trial_params": null
245
  }
 
3
  "best_model_checkpoint": null,
4
  "epoch": 1.0,
5
  "eval_steps": 500,
6
+ "global_step": 268,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
+ "epoch": 0.018656716417910446,
13
+ "grad_norm": 0.169921875,
14
+ "learning_rate": 1.785714285714286e-05,
15
+ "loss": 0.1062,
16
  "step": 5
17
  },
18
  {
19
+ "epoch": 0.03731343283582089,
20
+ "grad_norm": 0.126953125,
21
+ "learning_rate": 3.571428571428572e-05,
22
+ "loss": 0.0978,
23
  "step": 10
24
  },
25
  {
26
+ "epoch": 0.055970149253731345,
27
+ "grad_norm": 0.07666015625,
28
+ "learning_rate": 4.999827900623038e-05,
29
+ "loss": 0.0911,
30
  "step": 15
31
  },
32
  {
33
+ "epoch": 0.07462686567164178,
34
+ "grad_norm": 0.05322265625,
35
+ "learning_rate": 4.993807186343243e-05,
36
+ "loss": 0.0873,
37
  "step": 20
38
  },
39
  {
40
+ "epoch": 0.09328358208955224,
41
+ "grad_norm": 0.039794921875,
42
+ "learning_rate": 4.979207812402531e-05,
43
+ "loss": 0.0838,
44
  "step": 25
45
  },
46
  {
47
+ "epoch": 0.11194029850746269,
48
+ "grad_norm": 0.030029296875,
49
+ "learning_rate": 4.956085596012407e-05,
50
+ "loss": 0.0832,
51
  "step": 30
52
  },
53
  {
54
+ "epoch": 0.13059701492537312,
55
+ "grad_norm": 0.0252685546875,
56
+ "learning_rate": 4.924528939432311e-05,
57
+ "loss": 0.0814,
58
  "step": 35
59
  },
60
  {
61
+ "epoch": 0.14925373134328357,
62
+ "grad_norm": 0.021484375,
63
+ "learning_rate": 4.884658491984735e-05,
64
+ "loss": 0.0806,
65
  "step": 40
66
  },
67
  {
68
+ "epoch": 0.16791044776119404,
69
+ "grad_norm": 0.020751953125,
70
+ "learning_rate": 4.8366266887814235e-05,
71
+ "loss": 0.0809,
72
  "step": 45
73
  },
74
  {
75
+ "epoch": 0.1865671641791045,
76
+ "grad_norm": 0.01953125,
77
+ "learning_rate": 4.780617167924209e-05,
78
+ "loss": 0.0788,
79
  "step": 50
80
  },
81
  {
82
+ "epoch": 0.20522388059701493,
83
+ "grad_norm": 0.0205078125,
84
+ "learning_rate": 4.716844068408693e-05,
85
+ "loss": 0.0793,
86
  "step": 55
87
  },
88
  {
89
+ "epoch": 0.22388059701492538,
90
+ "grad_norm": 0.019775390625,
91
+ "learning_rate": 4.6455512114150546e-05,
92
+ "loss": 0.0789,
93
  "step": 60
94
  },
95
  {
96
+ "epoch": 0.24253731343283583,
97
+ "grad_norm": 0.01904296875,
98
+ "learning_rate": 4.5670111681161296e-05,
99
+ "loss": 0.0778,
100
  "step": 65
101
  },
102
  {
103
+ "epoch": 0.26119402985074625,
104
+ "grad_norm": 0.01953125,
105
+ "learning_rate": 4.481524217566783e-05,
106
+ "loss": 0.0775,
107
  "step": 70
108
  },
109
  {
110
+ "epoch": 0.2798507462686567,
111
+ "grad_norm": 0.01953125,
112
+ "learning_rate": 4.3894171986588217e-05,
113
+ "loss": 0.0773,
114
  "step": 75
115
  },
116
  {
117
+ "epoch": 0.29850746268656714,
118
+ "grad_norm": 0.0194091796875,
119
+ "learning_rate": 4.29104226053073e-05,
120
+ "loss": 0.0773,
121
  "step": 80
122
  },
123
  {
124
+ "epoch": 0.31716417910447764,
125
+ "grad_norm": 0.0184326171875,
126
+ "learning_rate": 4.186775516209732e-05,
127
+ "loss": 0.0768,
128
  "step": 85
129
  },
130
  {
131
+ "epoch": 0.3358208955223881,
132
+ "grad_norm": 0.0194091796875,
133
+ "learning_rate": 4.077015604633669e-05,
134
+ "loss": 0.0773,
135
  "step": 90
136
  },
137
  {
138
+ "epoch": 0.35447761194029853,
139
+ "grad_norm": 0.0191650390625,
140
+ "learning_rate": 3.962182166550441e-05,
141
+ "loss": 0.0772,
142
  "step": 95
143
  },
144
  {
145
+ "epoch": 0.373134328358209,
146
+ "grad_norm": 0.0185546875,
147
+ "learning_rate": 3.8427142401220634e-05,
148
+ "loss": 0.0768,
149
  "step": 100
150
  },
151
  {
152
+ "epoch": 0.3917910447761194,
153
+ "grad_norm": 0.0186767578125,
154
+ "learning_rate": 3.71906858236735e-05,
155
+ "loss": 0.0764,
156
  "step": 105
157
  },
158
  {
159
+ "epoch": 0.41044776119402987,
160
+ "grad_norm": 0.0179443359375,
161
+ "learning_rate": 3.591717922860785e-05,
162
+ "loss": 0.0773,
163
  "step": 110
164
  },
165
  {
166
+ "epoch": 0.4291044776119403,
167
+ "grad_norm": 0.017822265625,
168
+ "learning_rate": 3.46114915636416e-05,
169
+ "loss": 0.0762,
170
  "step": 115
171
  },
172
  {
173
+ "epoch": 0.44776119402985076,
174
+ "grad_norm": 0.0194091796875,
175
+ "learning_rate": 3.3278614813010034e-05,
176
+ "loss": 0.0765,
177
  "step": 120
178
  },
179
  {
180
+ "epoch": 0.4664179104477612,
181
+ "grad_norm": 0.0179443359375,
182
+ "learning_rate": 3.1923644911909e-05,
183
+ "loss": 0.0762,
184
  "step": 125
185
  },
186
  {
187
+ "epoch": 0.48507462686567165,
188
+ "grad_norm": 0.017578125,
189
+ "learning_rate": 3.0551762263406576e-05,
190
+ "loss": 0.0761,
191
  "step": 130
192
  },
193
  {
194
+ "epoch": 0.503731343283582,
195
+ "grad_norm": 0.0186767578125,
196
+ "learning_rate": 2.9168211932412042e-05,
197
+ "loss": 0.0758,
198
  "step": 135
199
  },
200
  {
201
+ "epoch": 0.5223880597014925,
202
+ "grad_norm": 0.0174560546875,
203
+ "learning_rate": 2.777828359242567e-05,
204
+ "loss": 0.0766,
205
  "step": 140
206
  },
207
  {
208
+ "epoch": 0.5410447761194029,
209
+ "grad_norm": 0.0179443359375,
210
+ "learning_rate": 2.6387291301738377e-05,
211
+ "loss": 0.0756,
212
  "step": 145
213
  },
214
+ {
215
+ "epoch": 0.5597014925373134,
216
+ "grad_norm": 0.01708984375,
217
+ "learning_rate": 2.50005531864019e-05,
218
+ "loss": 0.0755,
219
+ "step": 150
220
+ },
221
+ {
222
+ "epoch": 0.5783582089552238,
223
+ "grad_norm": 0.0169677734375,
224
+ "learning_rate": 2.362337110764688e-05,
225
+ "loss": 0.0755,
226
+ "step": 155
227
+ },
228
+ {
229
+ "epoch": 0.5970149253731343,
230
+ "grad_norm": 0.0181884765625,
231
+ "learning_rate": 2.226101039148557e-05,
232
+ "loss": 0.0754,
233
+ "step": 160
234
+ },
235
+ {
236
+ "epoch": 0.6156716417910447,
237
+ "grad_norm": 0.0169677734375,
238
+ "learning_rate": 2.0918679697998252e-05,
239
+ "loss": 0.0753,
240
+ "step": 165
241
+ },
242
+ {
243
+ "epoch": 0.6343283582089553,
244
+ "grad_norm": 0.017578125,
245
+ "learning_rate": 1.9601511107268255e-05,
246
+ "loss": 0.0755,
247
+ "step": 170
248
+ },
249
+ {
250
+ "epoch": 0.6529850746268657,
251
+ "grad_norm": 0.0181884765625,
252
+ "learning_rate": 1.8314540498102216e-05,
253
+ "loss": 0.0755,
254
+ "step": 175
255
+ },
256
+ {
257
+ "epoch": 0.6716417910447762,
258
+ "grad_norm": 0.016357421875,
259
+ "learning_rate": 1.7062688294552992e-05,
260
+ "loss": 0.0745,
261
+ "step": 180
262
+ },
263
+ {
264
+ "epoch": 0.6902985074626866,
265
+ "grad_norm": 0.0167236328125,
266
+ "learning_rate": 1.5850740653856096e-05,
267
+ "loss": 0.075,
268
+ "step": 185
269
+ },
270
+ {
271
+ "epoch": 0.7089552238805971,
272
+ "grad_norm": 0.0172119140625,
273
+ "learning_rate": 1.4683331167703218e-05,
274
+ "loss": 0.0756,
275
+ "step": 190
276
+ },
277
+ {
278
+ "epoch": 0.7276119402985075,
279
+ "grad_norm": 0.016845703125,
280
+ "learning_rate": 1.356492314681356e-05,
281
+ "loss": 0.076,
282
+ "step": 195
283
+ },
284
+ {
285
+ "epoch": 0.746268656716418,
286
+ "grad_norm": 0.0169677734375,
287
+ "learning_rate": 1.2499792556533716e-05,
288
+ "loss": 0.0753,
289
+ "step": 200
290
+ },
291
+ {
292
+ "epoch": 0.7649253731343284,
293
+ "grad_norm": 0.0167236328125,
294
+ "learning_rate": 1.1492011668707753e-05,
295
+ "loss": 0.075,
296
+ "step": 205
297
+ },
298
+ {
299
+ "epoch": 0.7835820895522388,
300
+ "grad_norm": 0.0169677734375,
301
+ "learning_rate": 1.0545433492320603e-05,
302
+ "loss": 0.0757,
303
+ "step": 210
304
+ },
305
+ {
306
+ "epoch": 0.8022388059701493,
307
+ "grad_norm": 0.0167236328125,
308
+ "learning_rate": 9.663677042440537e-06,
309
+ "loss": 0.075,
310
+ "step": 215
311
+ },
312
+ {
313
+ "epoch": 0.8208955223880597,
314
+ "grad_norm": 0.0167236328125,
315
+ "learning_rate": 8.850113503781367e-06,
316
+ "loss": 0.0751,
317
+ "step": 220
318
+ },
319
+ {
320
+ "epoch": 0.8395522388059702,
321
+ "grad_norm": 0.0167236328125,
322
+ "learning_rate": 8.107853341784671e-06,
323
+ "loss": 0.076,
324
+ "step": 225
325
+ },
326
+ {
327
+ "epoch": 0.8582089552238806,
328
+ "grad_norm": 0.01708984375,
329
+ "learning_rate": 7.439734410499752e-06,
330
+ "loss": 0.0755,
331
+ "step": 230
332
+ },
333
+ {
334
+ "epoch": 0.8768656716417911,
335
+ "grad_norm": 0.016357421875,
336
+ "learning_rate": 6.848311102728011e-06,
337
+ "loss": 0.0756,
338
+ "step": 235
339
+ },
340
+ {
341
+ "epoch": 0.8955223880597015,
342
+ "grad_norm": 0.0167236328125,
343
+ "learning_rate": 6.335844583913515e-06,
344
+ "loss": 0.0751,
345
+ "step": 240
346
+ },
347
+ {
348
+ "epoch": 0.914179104477612,
349
+ "grad_norm": 0.016357421875,
350
+ "learning_rate": 5.904294147118193e-06,
351
+ "loss": 0.0757,
352
+ "step": 245
353
+ },
354
+ {
355
+ "epoch": 0.9328358208955224,
356
+ "grad_norm": 0.0164794921875,
357
+ "learning_rate": 5.555309722133842e-06,
358
+ "loss": 0.0753,
359
+ "step": 250
360
+ },
361
+ {
362
+ "epoch": 0.9514925373134329,
363
+ "grad_norm": 0.0166015625,
364
+ "learning_rate": 5.290225567370509e-06,
365
+ "loss": 0.0747,
366
+ "step": 255
367
+ },
368
+ {
369
+ "epoch": 0.9701492537313433,
370
+ "grad_norm": 0.016845703125,
371
+ "learning_rate": 5.110055168638854e-06,
372
+ "loss": 0.0753,
373
+ "step": 260
374
+ },
375
+ {
376
+ "epoch": 0.9888059701492538,
377
+ "grad_norm": 0.016845703125,
378
+ "learning_rate": 5.0154873643297575e-06,
379
+ "loss": 0.0757,
380
+ "step": 265
381
+ },
382
  {
383
  "epoch": 1.0,
384
+ "step": 268,
385
+ "total_flos": 4.4183709150420664e+18,
386
+ "train_loss": 0.07813258930595952,
387
+ "train_runtime": 6469.3467,
388
+ "train_samples_per_second": 5.302,
389
+ "train_steps_per_second": 0.041
390
  }
391
  ],
392
  "logging_steps": 5,
393
+ "max_steps": 268,
394
  "num_input_tokens_seen": 0,
395
  "num_train_epochs": 1,
396
  "save_steps": 100,
 
406
  "attributes": {}
407
  }
408
  },
409
+ "total_flos": 4.4183709150420664e+18,
410
+ "train_batch_size": 128,
411
  "trial_name": null,
412
  "trial_params": null
413
  }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9078be8fad67d8ea2b345643aa6a0dcec2f78f5ee9c244b833374dd77f101649
3
- size 6136
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:44fea505bef14b3f613c107c7eb7d0f1b0186d9243be129ad6b4f1eec8a0e9ef
3
+ size 6072