Sunnylululu commited on
Commit
760bbec
·
verified ·
1 Parent(s): 720f504

Model save

Browse files
Files changed (5) hide show
  1. README.md +1 -1
  2. all_results.json +5 -5
  3. generation_config.json +10 -5
  4. train_results.json +5 -5
  5. trainer_state.json +588 -112
README.md CHANGED
@@ -26,7 +26,7 @@ print(output["generated_text"])
26
 
27
  ## Training procedure
28
 
29
- [<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/648848014/huggingface/runs/7eblkip7)
30
 
31
 
32
  This model was trained with SFT.
 
26
 
27
  ## Training procedure
28
 
29
+ [<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/648848014/huggingface/runs/bx5kbmeh)
30
 
31
 
32
  This model was trained with SFT.
all_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "total_flos": 2.926914460515809e+17,
3
- "train_loss": 0.0,
4
- "train_runtime": 1.5655,
5
  "train_samples": 3472,
6
- "train_samples_per_second": 1232.8,
7
- "train_steps_per_second": 83.038
8
  }
 
1
  {
2
+ "total_flos": 2.4184401017438208e+17,
3
+ "train_loss": 0.6382126904548483,
4
+ "train_runtime": 2326.1115,
5
  "train_samples": 3472,
6
+ "train_samples_per_second": 3.229,
7
+ "train_steps_per_second": 0.202
8
  }
generation_config.json CHANGED
@@ -1,9 +1,14 @@
1
  {
2
- "_from_model_config": true,
3
- "bos_token_id": 151646,
4
  "do_sample": true,
5
- "eos_token_id": 151643,
6
- "temperature": 0.6,
7
- "top_p": 0.95,
 
 
 
 
 
 
8
  "transformers_version": "4.50.0.dev0"
9
  }
 
1
  {
2
+ "bos_token_id": 151643,
 
3
  "do_sample": true,
4
+ "eos_token_id": [
5
+ 151645,
6
+ 151643
7
+ ],
8
+ "pad_token_id": 151643,
9
+ "repetition_penalty": 1.1,
10
+ "temperature": 0.7,
11
+ "top_k": 20,
12
+ "top_p": 0.8,
13
  "transformers_version": "4.50.0.dev0"
14
  }
train_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "total_flos": 2.926914460515809e+17,
3
- "train_loss": 0.0,
4
- "train_runtime": 1.5655,
5
  "train_samples": 3472,
6
- "train_samples_per_second": 1232.8,
7
- "train_steps_per_second": 83.038
8
  }
 
1
  {
2
+ "total_flos": 2.4184401017438208e+17,
3
+ "train_loss": 0.6382126904548483,
4
+ "train_runtime": 2326.1115,
5
  "train_samples": 3472,
6
+ "train_samples_per_second": 3.229,
7
+ "train_steps_per_second": 0.202
8
  }
trainer_state.json CHANGED
@@ -3,205 +3,681 @@
3
  "best_model_checkpoint": null,
4
  "epoch": 10.0,
5
  "eval_steps": 500,
6
- "global_step": 130,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
- "epoch": 0.38461538461538464,
13
- "grad_norm": 0.9453125,
14
- "learning_rate": 3.875e-05,
15
- "loss": 1.7199,
16
  "step": 5
17
  },
18
  {
19
- "epoch": 0.7692307692307693,
20
- "grad_norm": 0.5234375,
21
- "learning_rate": 1.1590097423302684e-05,
22
- "loss": 1.4892,
23
  "step": 10
24
  },
25
  {
26
- "epoch": 1.1538461538461537,
27
- "grad_norm": 0.478515625,
28
- "learning_rate": 4.953193036870676e-05,
29
- "loss": 1.4358,
30
  "step": 15
31
  },
32
  {
33
- "epoch": 1.5384615384615383,
34
- "grad_norm": 0.4921875,
35
- "learning_rate": 4.877104772313846e-05,
36
- "loss": 1.4053,
37
  "step": 20
38
  },
39
  {
40
- "epoch": 1.9230769230769231,
41
- "grad_norm": 0.40234375,
42
- "learning_rate": 4.766372503162375e-05,
43
- "loss": 1.3156,
44
  "step": 25
45
  },
46
  {
47
- "epoch": 2.3076923076923075,
48
- "grad_norm": 0.380859375,
49
- "learning_rate": 4.622799718031961e-05,
50
- "loss": 1.2542,
51
  "step": 30
52
  },
53
  {
54
- "epoch": 2.6923076923076925,
55
- "grad_norm": 0.341796875,
56
- "learning_rate": 4.448724776693342e-05,
57
- "loss": 1.2351,
58
  "step": 35
59
  },
60
  {
61
- "epoch": 3.076923076923077,
62
- "grad_norm": 0.318359375,
63
- "learning_rate": 4.246982825372522e-05,
64
- "loss": 1.2441,
65
  "step": 40
66
  },
67
  {
68
- "epoch": 3.4615384615384617,
69
- "grad_norm": 0.31640625,
70
- "learning_rate": 4.020859620925235e-05,
71
- "loss": 1.1722,
72
  "step": 45
73
  },
74
  {
75
- "epoch": 3.8461538461538463,
76
- "grad_norm": 0.302734375,
77
- "learning_rate": 3.77403801594802e-05,
78
- "loss": 1.1937,
79
  "step": 50
80
  },
81
  {
82
- "epoch": 4.230769230769231,
83
- "grad_norm": 0.298828125,
84
- "learning_rate": 3.510537976419131e-05,
85
- "loss": 1.1898,
86
  "step": 55
87
  },
88
  {
89
- "epoch": 4.615384615384615,
90
- "grad_norm": 0.306640625,
91
- "learning_rate": 3.234651108797708e-05,
92
- "loss": 1.1526,
93
  "step": 60
94
  },
95
  {
96
- "epoch": 5.0,
97
- "grad_norm": 1.0703125,
98
- "learning_rate": 2.9508707629336874e-05,
99
- "loss": 1.1345,
100
  "step": 65
101
  },
102
  {
103
- "epoch": 5.384615384615385,
104
- "grad_norm": 0.2890625,
105
- "learning_rate": 2.6638188491974207e-05,
106
- "loss": 1.1493,
107
  "step": 70
108
  },
109
  {
110
- "epoch": 5.769230769230769,
111
- "grad_norm": 0.28515625,
112
- "learning_rate": 2.378170561753209e-05,
113
- "loss": 1.121,
114
  "step": 75
115
  },
116
  {
117
- "epoch": 6.153846153846154,
118
- "grad_norm": 0.2734375,
119
- "learning_rate": 2.098578234003466e-05,
120
- "loss": 1.1091,
121
  "step": 80
122
  },
123
  {
124
- "epoch": 6.538461538461538,
125
- "grad_norm": 0.2734375,
126
- "learning_rate": 1.8295955663644855e-05,
127
- "loss": 1.1314,
128
  "step": 85
129
  },
130
  {
131
- "epoch": 6.923076923076923,
132
- "grad_norm": 0.271484375,
133
- "learning_rate": 1.575603460470665e-05,
134
- "loss": 1.1291,
135
  "step": 90
136
  },
137
  {
138
- "epoch": 7.3076923076923075,
139
- "grad_norm": 0.271484375,
140
- "learning_rate": 1.3407386677402312e-05,
141
- "loss": 1.123,
142
  "step": 95
143
  },
144
  {
145
- "epoch": 7.6923076923076925,
146
- "grad_norm": 0.28125,
147
- "learning_rate": 1.1288264143982296e-05,
148
- "loss": 1.1186,
149
  "step": 100
150
  },
151
  {
152
- "epoch": 8.076923076923077,
153
- "grad_norm": 0.259765625,
154
- "learning_rate": 9.433181002882383e-06,
155
- "loss": 1.096,
156
  "step": 105
157
  },
158
  {
159
- "epoch": 8.461538461538462,
160
- "grad_norm": 0.275390625,
161
- "learning_rate": 7.872350861678565e-06,
162
- "loss": 1.1196,
163
  "step": 110
164
  },
165
  {
166
- "epoch": 8.846153846153847,
167
- "grad_norm": 0.255859375,
168
- "learning_rate": 6.631194850202872e-06,
169
- "loss": 1.1171,
170
  "step": 115
171
  },
172
  {
173
- "epoch": 9.23076923076923,
174
- "grad_norm": 0.283203125,
175
- "learning_rate": 5.729927588404395e-06,
176
- "loss": 1.1002,
177
  "step": 120
178
  },
179
  {
180
- "epoch": 9.615384615384615,
181
- "grad_norm": 0.259765625,
182
- "learning_rate": 5.1832279522675925e-06,
183
- "loss": 1.0983,
184
  "step": 125
185
  },
186
  {
187
- "epoch": 10.0,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
188
  "grad_norm": 0.96484375,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
189
  "learning_rate": 5e-06,
190
- "loss": 1.095,
191
- "step": 130
192
  },
193
  {
194
  "epoch": 10.0,
195
- "step": 130,
196
- "total_flos": 2.926914460515809e+17,
197
- "train_loss": 0.0,
198
- "train_runtime": 1.5655,
199
- "train_samples_per_second": 1232.8,
200
- "train_steps_per_second": 83.038
201
  }
202
  ],
203
  "logging_steps": 5,
204
- "max_steps": 130,
205
  "num_input_tokens_seen": 0,
206
  "num_train_epochs": 10,
207
  "save_steps": 100,
@@ -217,7 +693,7 @@
217
  "attributes": {}
218
  }
219
  },
220
- "total_flos": 2.926914460515809e+17,
221
  "train_batch_size": 16,
222
  "trial_name": null,
223
  "trial_params": null
 
3
  "best_model_checkpoint": null,
4
  "epoch": 10.0,
5
  "eval_steps": 500,
6
+ "global_step": 470,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
+ "epoch": 0.10638297872340426,
13
+ "grad_norm": 3.921875,
14
+ "learning_rate": 1.0416666666666668e-05,
15
+ "loss": 1.6159,
16
  "step": 5
17
  },
18
  {
19
+ "epoch": 0.2127659574468085,
20
+ "grad_norm": 1.796875,
21
+ "learning_rate": 2.0833333333333336e-05,
22
+ "loss": 1.4989,
23
  "step": 10
24
  },
25
  {
26
+ "epoch": 0.3191489361702128,
27
+ "grad_norm": 1.4609375,
28
+ "learning_rate": 3.125e-05,
29
+ "loss": 1.3278,
30
  "step": 15
31
  },
32
  {
33
+ "epoch": 0.425531914893617,
34
+ "grad_norm": 1.2421875,
35
+ "learning_rate": 4.166666666666667e-05,
36
+ "loss": 1.2225,
37
  "step": 20
38
  },
39
  {
40
+ "epoch": 0.5319148936170213,
41
+ "grad_norm": 1.2109375,
42
+ "learning_rate": 4.999944181166713e-05,
43
+ "loss": 1.1515,
44
  "step": 25
45
  },
46
  {
47
+ "epoch": 0.6382978723404256,
48
+ "grad_norm": 1.03125,
49
+ "learning_rate": 4.997990812788907e-05,
50
+ "loss": 1.1002,
51
  "step": 30
52
  },
53
  {
54
+ "epoch": 0.7446808510638298,
55
+ "grad_norm": 1.0390625,
56
+ "learning_rate": 4.993249271677745e-05,
57
+ "loss": 1.0933,
58
  "step": 35
59
  },
60
  {
61
+ "epoch": 0.851063829787234,
62
+ "grad_norm": 1.046875,
63
+ "learning_rate": 4.985725438745002e-05,
64
+ "loss": 1.0604,
65
  "step": 40
66
  },
67
  {
68
+ "epoch": 0.9574468085106383,
69
+ "grad_norm": 1.0546875,
70
+ "learning_rate": 4.975428645766288e-05,
71
+ "loss": 1.0401,
72
  "step": 45
73
  },
74
  {
75
+ "epoch": 1.0638297872340425,
76
+ "grad_norm": 1.0078125,
77
+ "learning_rate": 4.962371663806883e-05,
78
+ "loss": 0.9942,
79
  "step": 50
80
  },
81
  {
82
+ "epoch": 1.1702127659574468,
83
+ "grad_norm": 1.1171875,
84
+ "learning_rate": 4.9465706873818514e-05,
85
+ "loss": 0.9376,
86
  "step": 55
87
  },
88
  {
89
+ "epoch": 1.2765957446808511,
90
+ "grad_norm": 1.0390625,
91
+ "learning_rate": 4.928045314370053e-05,
92
+ "loss": 0.9253,
93
  "step": 60
94
  },
95
  {
96
+ "epoch": 1.3829787234042552,
97
+ "grad_norm": 1.0234375,
98
+ "learning_rate": 4.906818521706987e-05,
99
+ "loss": 0.9234,
100
  "step": 65
101
  },
102
  {
103
+ "epoch": 1.4893617021276595,
104
+ "grad_norm": 0.9375,
105
+ "learning_rate": 4.8829166368866006e-05,
106
+ "loss": 0.925,
107
  "step": 70
108
  },
109
  {
110
+ "epoch": 1.5957446808510638,
111
+ "grad_norm": 0.96484375,
112
+ "learning_rate": 4.8563693053074235e-05,
113
+ "loss": 0.9129,
114
  "step": 75
115
  },
116
  {
117
+ "epoch": 1.702127659574468,
118
+ "grad_norm": 0.97265625,
119
+ "learning_rate": 4.8272094535035095e-05,
120
+ "loss": 0.8839,
121
  "step": 80
122
  },
123
  {
124
+ "epoch": 1.8085106382978724,
125
+ "grad_norm": 0.9140625,
126
+ "learning_rate": 4.7954732483058103e-05,
127
+ "loss": 0.8724,
128
  "step": 85
129
  },
130
  {
131
+ "epoch": 1.9148936170212765,
132
+ "grad_norm": 1.0859375,
133
+ "learning_rate": 4.761200051984621e-05,
134
+ "loss": 0.8852,
135
  "step": 90
136
  },
137
  {
138
+ "epoch": 2.021276595744681,
139
+ "grad_norm": 1.046875,
140
+ "learning_rate": 4.724432373428734e-05,
141
+ "loss": 0.8763,
142
  "step": 95
143
  },
144
  {
145
+ "epoch": 2.127659574468085,
146
+ "grad_norm": 1.0390625,
147
+ "learning_rate": 4.6852158154218594e-05,
148
+ "loss": 0.7757,
149
  "step": 100
150
  },
151
  {
152
+ "epoch": 2.2340425531914896,
153
+ "grad_norm": 1.1015625,
154
+ "learning_rate": 4.643599018081702e-05,
155
+ "loss": 0.7828,
156
  "step": 105
157
  },
158
  {
159
+ "epoch": 2.3404255319148937,
160
+ "grad_norm": 1.109375,
161
+ "learning_rate": 4.5996335985318416e-05,
162
+ "loss": 0.794,
163
  "step": 110
164
  },
165
  {
166
+ "epoch": 2.4468085106382977,
167
+ "grad_norm": 1.03125,
168
+ "learning_rate": 4.553374086881255e-05,
169
+ "loss": 0.7854,
170
  "step": 115
171
  },
172
  {
173
+ "epoch": 2.5531914893617023,
174
+ "grad_norm": 1.140625,
175
+ "learning_rate": 4.5048778585908696e-05,
176
+ "loss": 0.7727,
177
  "step": 120
178
  },
179
  {
180
+ "epoch": 2.6595744680851063,
181
+ "grad_norm": 1.1015625,
182
+ "learning_rate": 4.454205063311039e-05,
183
+ "loss": 0.7748,
184
  "step": 125
185
  },
186
  {
187
+ "epoch": 2.7659574468085104,
188
+ "grad_norm": 0.94921875,
189
+ "learning_rate": 4.401418550278211e-05,
190
+ "loss": 0.7644,
191
+ "step": 130
192
+ },
193
+ {
194
+ "epoch": 2.872340425531915,
195
+ "grad_norm": 1.140625,
196
+ "learning_rate": 4.346583790363301e-05,
197
+ "loss": 0.7659,
198
+ "step": 135
199
+ },
200
+ {
201
+ "epoch": 2.978723404255319,
202
+ "grad_norm": 1.0078125,
203
+ "learning_rate": 4.289768794868476e-05,
204
+ "loss": 0.7569,
205
+ "step": 140
206
+ },
207
+ {
208
+ "epoch": 3.0851063829787235,
209
+ "grad_norm": 1.1015625,
210
+ "learning_rate": 4.231044031173043e-05,
211
+ "loss": 0.6841,
212
+ "step": 145
213
+ },
214
+ {
215
+ "epoch": 3.1914893617021276,
216
+ "grad_norm": 1.0078125,
217
+ "learning_rate": 4.170482335333083e-05,
218
+ "loss": 0.6804,
219
+ "step": 150
220
+ },
221
+ {
222
+ "epoch": 3.297872340425532,
223
+ "grad_norm": 1.09375,
224
+ "learning_rate": 4.10815882174322e-05,
225
+ "loss": 0.6608,
226
+ "step": 155
227
+ },
228
+ {
229
+ "epoch": 3.404255319148936,
230
+ "grad_norm": 1.078125,
231
+ "learning_rate": 4.0441507899725865e-05,
232
+ "loss": 0.6677,
233
+ "step": 160
234
+ },
235
+ {
236
+ "epoch": 3.5106382978723403,
237
+ "grad_norm": 0.96484375,
238
+ "learning_rate": 3.978537628890514e-05,
239
+ "loss": 0.6568,
240
+ "step": 165
241
+ },
242
+ {
243
+ "epoch": 3.617021276595745,
244
+ "grad_norm": 1.09375,
245
+ "learning_rate": 3.9114007182008886e-05,
246
+ "loss": 0.6551,
247
+ "step": 170
248
+ },
249
+ {
250
+ "epoch": 3.723404255319149,
251
+ "grad_norm": 1.078125,
252
+ "learning_rate": 3.842823327507267e-05,
253
+ "loss": 0.6403,
254
+ "step": 175
255
+ },
256
+ {
257
+ "epoch": 3.829787234042553,
258
+ "grad_norm": 1.0390625,
259
+ "learning_rate": 3.772890513033978e-05,
260
+ "loss": 0.673,
261
+ "step": 180
262
+ },
263
+ {
264
+ "epoch": 3.9361702127659575,
265
+ "grad_norm": 1.0859375,
266
+ "learning_rate": 3.701689012131268e-05,
267
+ "loss": 0.6611,
268
+ "step": 185
269
+ },
270
+ {
271
+ "epoch": 4.042553191489362,
272
+ "grad_norm": 1.015625,
273
+ "learning_rate": 3.629307135695365e-05,
274
+ "loss": 0.6185,
275
+ "step": 190
276
+ },
277
+ {
278
+ "epoch": 4.148936170212766,
279
+ "grad_norm": 1.0078125,
280
+ "learning_rate": 3.555834658636878e-05,
281
+ "loss": 0.5792,
282
+ "step": 195
283
+ },
284
+ {
285
+ "epoch": 4.25531914893617,
286
+ "grad_norm": 1.140625,
287
+ "learning_rate": 3.4813627085333855e-05,
288
+ "loss": 0.556,
289
+ "step": 200
290
+ },
291
+ {
292
+ "epoch": 4.361702127659575,
293
+ "grad_norm": 1.0546875,
294
+ "learning_rate": 3.405983652604315e-05,
295
+ "loss": 0.5776,
296
+ "step": 205
297
+ },
298
+ {
299
+ "epoch": 4.468085106382979,
300
+ "grad_norm": 1.1015625,
301
+ "learning_rate": 3.3297909831483065e-05,
302
+ "loss": 0.573,
303
+ "step": 210
304
+ },
305
+ {
306
+ "epoch": 4.574468085106383,
307
+ "grad_norm": 1.1875,
308
+ "learning_rate": 3.252879201585139e-05,
309
+ "loss": 0.5789,
310
+ "step": 215
311
+ },
312
+ {
313
+ "epoch": 4.680851063829787,
314
+ "grad_norm": 1.0546875,
315
+ "learning_rate": 3.175343701246051e-05,
316
+ "loss": 0.574,
317
+ "step": 220
318
+ },
319
+ {
320
+ "epoch": 4.787234042553192,
321
+ "grad_norm": 1.046875,
322
+ "learning_rate": 3.097280649057833e-05,
323
+ "loss": 0.5644,
324
+ "step": 225
325
+ },
326
+ {
327
+ "epoch": 4.8936170212765955,
328
+ "grad_norm": 1.03125,
329
+ "learning_rate": 3.0187868662674168e-05,
330
+ "loss": 0.5646,
331
+ "step": 230
332
+ },
333
+ {
334
+ "epoch": 5.0,
335
+ "grad_norm": 0.98828125,
336
+ "learning_rate": 2.9399597083549386e-05,
337
+ "loss": 0.566,
338
+ "step": 235
339
+ },
340
+ {
341
+ "epoch": 5.1063829787234045,
342
+ "grad_norm": 1.1328125,
343
+ "learning_rate": 2.8608969442841698e-05,
344
+ "loss": 0.5118,
345
+ "step": 240
346
+ },
347
+ {
348
+ "epoch": 5.212765957446808,
349
+ "grad_norm": 1.0078125,
350
+ "learning_rate": 2.781696635240128e-05,
351
+ "loss": 0.5009,
352
+ "step": 245
353
+ },
354
+ {
355
+ "epoch": 5.319148936170213,
356
+ "grad_norm": 1.0390625,
357
+ "learning_rate": 2.702457013004228e-05,
358
+ "loss": 0.5057,
359
+ "step": 250
360
+ },
361
+ {
362
+ "epoch": 5.425531914893617,
363
+ "grad_norm": 1.0546875,
364
+ "learning_rate": 2.623276358117861e-05,
365
+ "loss": 0.5049,
366
+ "step": 255
367
+ },
368
+ {
369
+ "epoch": 5.531914893617021,
370
+ "grad_norm": 1.0234375,
371
+ "learning_rate": 2.5442528779854824e-05,
372
+ "loss": 0.511,
373
+ "step": 260
374
+ },
375
+ {
376
+ "epoch": 5.638297872340425,
377
+ "grad_norm": 1.0,
378
+ "learning_rate": 2.4654845850684177e-05,
379
+ "loss": 0.4898,
380
+ "step": 265
381
+ },
382
+ {
383
+ "epoch": 5.74468085106383,
384
+ "grad_norm": 0.9921875,
385
+ "learning_rate": 2.3870691753204553e-05,
386
+ "loss": 0.5094,
387
+ "step": 270
388
+ },
389
+ {
390
+ "epoch": 5.851063829787234,
391
+ "grad_norm": 0.9921875,
392
+ "learning_rate": 2.3091039070160044e-05,
393
+ "loss": 0.5061,
394
+ "step": 275
395
+ },
396
+ {
397
+ "epoch": 5.957446808510638,
398
+ "grad_norm": 1.0,
399
+ "learning_rate": 2.2316854801210935e-05,
400
+ "loss": 0.4936,
401
+ "step": 280
402
+ },
403
+ {
404
+ "epoch": 6.0638297872340425,
405
+ "grad_norm": 0.97265625,
406
+ "learning_rate": 2.1549099163568527e-05,
407
+ "loss": 0.4771,
408
+ "step": 285
409
+ },
410
+ {
411
+ "epoch": 6.170212765957447,
412
+ "grad_norm": 0.98046875,
413
+ "learning_rate": 2.0788724401042096e-05,
414
+ "loss": 0.4644,
415
+ "step": 290
416
+ },
417
+ {
418
+ "epoch": 6.276595744680851,
419
+ "grad_norm": 1.0,
420
+ "learning_rate": 2.0036673602975203e-05,
421
+ "loss": 0.4519,
422
+ "step": 295
423
+ },
424
+ {
425
+ "epoch": 6.382978723404255,
426
  "grad_norm": 0.96484375,
427
+ "learning_rate": 1.9293879534536358e-05,
428
+ "loss": 0.4626,
429
+ "step": 300
430
+ },
431
+ {
432
+ "epoch": 6.48936170212766,
433
+ "grad_norm": 0.9765625,
434
+ "learning_rate": 1.8561263479814583e-05,
435
+ "loss": 0.4594,
436
+ "step": 305
437
+ },
438
+ {
439
+ "epoch": 6.595744680851064,
440
+ "grad_norm": 0.98046875,
441
+ "learning_rate": 1.783973409915503e-05,
442
+ "loss": 0.4575,
443
+ "step": 310
444
+ },
445
+ {
446
+ "epoch": 6.702127659574468,
447
+ "grad_norm": 0.97265625,
448
+ "learning_rate": 1.713018630215162e-05,
449
+ "loss": 0.4595,
450
+ "step": 315
451
+ },
452
+ {
453
+ "epoch": 6.808510638297872,
454
+ "grad_norm": 0.984375,
455
+ "learning_rate": 1.6433500137694846e-05,
456
+ "loss": 0.461,
457
+ "step": 320
458
+ },
459
+ {
460
+ "epoch": 6.914893617021277,
461
+ "grad_norm": 0.98828125,
462
+ "learning_rate": 1.5750539702451116e-05,
463
+ "loss": 0.4578,
464
+ "step": 325
465
+ },
466
+ {
467
+ "epoch": 7.0212765957446805,
468
+ "grad_norm": 0.9921875,
469
+ "learning_rate": 1.5082152069127619e-05,
470
+ "loss": 0.4584,
471
+ "step": 330
472
+ },
473
+ {
474
+ "epoch": 7.127659574468085,
475
+ "grad_norm": 0.94921875,
476
+ "learning_rate": 1.442916623585189e-05,
477
+ "loss": 0.4262,
478
+ "step": 335
479
+ },
480
+ {
481
+ "epoch": 7.23404255319149,
482
+ "grad_norm": 0.95703125,
483
+ "learning_rate": 1.3792392097969259e-05,
484
+ "loss": 0.4352,
485
+ "step": 340
486
+ },
487
+ {
488
+ "epoch": 7.340425531914893,
489
+ "grad_norm": 0.9375,
490
+ "learning_rate": 1.3172619443533333e-05,
491
+ "loss": 0.4332,
492
+ "step": 345
493
+ },
494
+ {
495
+ "epoch": 7.446808510638298,
496
+ "grad_norm": 0.94140625,
497
+ "learning_rate": 1.2570616973735466e-05,
498
+ "loss": 0.4407,
499
+ "step": 350
500
+ },
501
+ {
502
+ "epoch": 7.553191489361702,
503
+ "grad_norm": 0.97265625,
504
+ "learning_rate": 1.1987131349488227e-05,
505
+ "loss": 0.4204,
506
+ "step": 355
507
+ },
508
+ {
509
+ "epoch": 7.659574468085106,
510
+ "grad_norm": 0.98046875,
511
+ "learning_rate": 1.1422886265345257e-05,
512
+ "loss": 0.4319,
513
+ "step": 360
514
+ },
515
+ {
516
+ "epoch": 7.76595744680851,
517
+ "grad_norm": 0.98828125,
518
+ "learning_rate": 1.0878581551906195e-05,
519
+ "loss": 0.4308,
520
+ "step": 365
521
+ },
522
+ {
523
+ "epoch": 7.872340425531915,
524
+ "grad_norm": 0.921875,
525
+ "learning_rate": 1.0354892307820033e-05,
526
+ "loss": 0.4462,
527
+ "step": 370
528
+ },
529
+ {
530
+ "epoch": 7.9787234042553195,
531
+ "grad_norm": 0.94921875,
532
+ "learning_rate": 9.852468062463322e-06,
533
+ "loss": 0.4336,
534
+ "step": 375
535
+ },
536
+ {
537
+ "epoch": 8.085106382978724,
538
+ "grad_norm": 0.96875,
539
+ "learning_rate": 9.371931970331851e-06,
540
+ "loss": 0.4244,
541
+ "step": 380
542
+ },
543
+ {
544
+ "epoch": 8.191489361702128,
545
+ "grad_norm": 0.953125,
546
+ "learning_rate": 8.913880038144973e-06,
547
+ "loss": 0.4271,
548
+ "step": 385
549
+ },
550
+ {
551
+ "epoch": 8.297872340425531,
552
+ "grad_norm": 0.96484375,
553
+ "learning_rate": 8.478880385621198e-06,
554
+ "loss": 0.4232,
555
+ "step": 390
556
+ },
557
+ {
558
+ "epoch": 8.404255319148936,
559
+ "grad_norm": 0.921875,
560
+ "learning_rate": 8.067472540841918e-06,
561
+ "loss": 0.4183,
562
+ "step": 395
563
+ },
564
+ {
565
+ "epoch": 8.51063829787234,
566
+ "grad_norm": 0.95703125,
567
+ "learning_rate": 7.680166771077155e-06,
568
+ "loss": 0.4184,
569
+ "step": 400
570
+ },
571
+ {
572
+ "epoch": 8.617021276595745,
573
+ "grad_norm": 0.9296875,
574
+ "learning_rate": 7.317443449903435e-06,
575
+ "loss": 0.4267,
576
+ "step": 405
577
+ },
578
+ {
579
+ "epoch": 8.72340425531915,
580
+ "grad_norm": 0.95703125,
581
+ "learning_rate": 6.9797524613986255e-06,
582
+ "loss": 0.4276,
583
+ "step": 410
584
+ },
585
+ {
586
+ "epoch": 8.829787234042554,
587
+ "grad_norm": 0.9453125,
588
+ "learning_rate": 6.667512642152743e-06,
589
+ "loss": 0.4178,
590
+ "step": 415
591
+ },
592
+ {
593
+ "epoch": 8.936170212765958,
594
+ "grad_norm": 0.95703125,
595
+ "learning_rate": 6.381111261786892e-06,
596
+ "loss": 0.4223,
597
+ "step": 420
598
+ },
599
+ {
600
+ "epoch": 9.042553191489361,
601
+ "grad_norm": 0.890625,
602
+ "learning_rate": 6.120903542624479e-06,
603
+ "loss": 0.4238,
604
+ "step": 425
605
+ },
606
+ {
607
+ "epoch": 9.148936170212766,
608
+ "grad_norm": 0.9296875,
609
+ "learning_rate": 5.887212219110616e-06,
610
+ "loss": 0.4208,
611
+ "step": 430
612
+ },
613
+ {
614
+ "epoch": 9.25531914893617,
615
+ "grad_norm": 0.921875,
616
+ "learning_rate": 5.6803271375260525e-06,
617
+ "loss": 0.4079,
618
+ "step": 435
619
+ },
620
+ {
621
+ "epoch": 9.361702127659575,
622
+ "grad_norm": 0.93359375,
623
+ "learning_rate": 5.500504896492159e-06,
624
+ "loss": 0.4287,
625
+ "step": 440
626
+ },
627
+ {
628
+ "epoch": 9.46808510638298,
629
+ "grad_norm": 0.9453125,
630
+ "learning_rate": 5.347968528712824e-06,
631
+ "loss": 0.426,
632
+ "step": 445
633
+ },
634
+ {
635
+ "epoch": 9.574468085106384,
636
+ "grad_norm": 0.97265625,
637
+ "learning_rate": 5.222907224348022e-06,
638
+ "loss": 0.4077,
639
+ "step": 450
640
+ },
641
+ {
642
+ "epoch": 9.680851063829786,
643
+ "grad_norm": 0.8984375,
644
+ "learning_rate": 5.125476096362112e-06,
645
+ "loss": 0.4208,
646
+ "step": 455
647
+ },
648
+ {
649
+ "epoch": 9.787234042553191,
650
+ "grad_norm": 0.94140625,
651
+ "learning_rate": 5.055795988137946e-06,
652
+ "loss": 0.4104,
653
+ "step": 460
654
+ },
655
+ {
656
+ "epoch": 9.893617021276595,
657
+ "grad_norm": 0.93359375,
658
+ "learning_rate": 5.013953323595361e-06,
659
+ "loss": 0.4249,
660
+ "step": 465
661
+ },
662
+ {
663
+ "epoch": 10.0,
664
+ "grad_norm": 0.94921875,
665
  "learning_rate": 5e-06,
666
+ "loss": 0.4235,
667
+ "step": 470
668
  },
669
  {
670
  "epoch": 10.0,
671
+ "step": 470,
672
+ "total_flos": 2.4184401017438208e+17,
673
+ "train_loss": 0.6382126904548483,
674
+ "train_runtime": 2326.1115,
675
+ "train_samples_per_second": 3.229,
676
+ "train_steps_per_second": 0.202
677
  }
678
  ],
679
  "logging_steps": 5,
680
+ "max_steps": 470,
681
  "num_input_tokens_seen": 0,
682
  "num_train_epochs": 10,
683
  "save_steps": 100,
 
693
  "attributes": {}
694
  }
695
  },
696
+ "total_flos": 2.4184401017438208e+17,
697
  "train_batch_size": 16,
698
  "trial_name": null,
699
  "trial_params": null