Commit
•
f842923
1
Parent(s):
33fafd4
End of training
Browse files- all_results.json +12 -6
- eval_results.json +7 -6
- train_results.json +8 -0
- trainer_state.json +265 -0
all_results.json
CHANGED
|
@@ -1,9 +1,15 @@
|
|
| 1 |
{
|
| 2 |
-
"
|
| 3 |
-
"
|
| 4 |
-
"
|
|
|
|
| 5 |
"eval_samples": 100,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
-
"perplexity":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_accuracy": 0.040325203252032524,
|
| 4 |
+
"eval_loss": 2.6953125,
|
| 5 |
+
"eval_runtime": 5.8054,
|
| 6 |
"eval_samples": 100,
|
| 7 |
+
"eval_samples_per_second": 17.225,
|
| 8 |
+
"eval_steps_per_second": 0.345,
|
| 9 |
+
"perplexity": 14.81014620089916,
|
| 10 |
+
"train_loss": 2.8766326904296875,
|
| 11 |
+
"train_runtime": 267.0978,
|
| 12 |
+
"train_samples": 1000,
|
| 13 |
+
"train_samples_per_second": 3.744,
|
| 14 |
+
"train_steps_per_second": 0.06
|
| 15 |
}
|
eval_results.json
CHANGED
|
@@ -1,9 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"
|
| 3 |
-
"
|
| 4 |
-
"
|
|
|
|
| 5 |
"eval_samples": 100,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
-
"perplexity":
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_accuracy": 0.040325203252032524,
|
| 4 |
+
"eval_loss": 2.6953125,
|
| 5 |
+
"eval_runtime": 5.8054,
|
| 6 |
"eval_samples": 100,
|
| 7 |
+
"eval_samples_per_second": 17.225,
|
| 8 |
+
"eval_steps_per_second": 0.345,
|
| 9 |
+
"perplexity": 14.81014620089916
|
| 10 |
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 2.8766326904296875,
|
| 4 |
+
"train_runtime": 267.0978,
|
| 5 |
+
"train_samples": 1000,
|
| 6 |
+
"train_samples_per_second": 3.744,
|
| 7 |
+
"train_steps_per_second": 0.06
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,265 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"global_step": 16,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.06,
|
| 12 |
+
"learning_rate": 0,
|
| 13 |
+
"loss": 2.9912,
|
| 14 |
+
"step": 1
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.06,
|
| 18 |
+
"eval_accuracy": 0.040365853658536587,
|
| 19 |
+
"eval_loss": 2.744140625,
|
| 20 |
+
"eval_runtime": 4.9375,
|
| 21 |
+
"eval_samples_per_second": 20.253,
|
| 22 |
+
"eval_steps_per_second": 0.405,
|
| 23 |
+
"step": 1
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.12,
|
| 27 |
+
"learning_rate": 0,
|
| 28 |
+
"loss": 2.9329,
|
| 29 |
+
"step": 2
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.12,
|
| 33 |
+
"eval_accuracy": 0.040365853658536587,
|
| 34 |
+
"eval_loss": 2.744140625,
|
| 35 |
+
"eval_runtime": 6.0216,
|
| 36 |
+
"eval_samples_per_second": 16.607,
|
| 37 |
+
"eval_steps_per_second": 0.332,
|
| 38 |
+
"step": 2
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.19,
|
| 42 |
+
"learning_rate": 0.0,
|
| 43 |
+
"loss": 2.9138,
|
| 44 |
+
"step": 3
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.19,
|
| 48 |
+
"eval_accuracy": 0.038902439024390244,
|
| 49 |
+
"eval_loss": 2.826171875,
|
| 50 |
+
"eval_runtime": 6.6113,
|
| 51 |
+
"eval_samples_per_second": 15.126,
|
| 52 |
+
"eval_steps_per_second": 0.303,
|
| 53 |
+
"step": 3
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.25,
|
| 57 |
+
"learning_rate": 5e-05,
|
| 58 |
+
"loss": 2.9395,
|
| 59 |
+
"step": 4
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.25,
|
| 63 |
+
"eval_accuracy": 0.038902439024390244,
|
| 64 |
+
"eval_loss": 2.826171875,
|
| 65 |
+
"eval_runtime": 7.2265,
|
| 66 |
+
"eval_samples_per_second": 13.838,
|
| 67 |
+
"eval_steps_per_second": 0.277,
|
| 68 |
+
"step": 4
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.31,
|
| 72 |
+
"learning_rate": 5e-05,
|
| 73 |
+
"loss": 2.9109,
|
| 74 |
+
"step": 5
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.31,
|
| 78 |
+
"eval_accuracy": 0.039878048780487806,
|
| 79 |
+
"eval_loss": 2.794921875,
|
| 80 |
+
"eval_runtime": 7.453,
|
| 81 |
+
"eval_samples_per_second": 13.417,
|
| 82 |
+
"eval_steps_per_second": 0.268,
|
| 83 |
+
"step": 5
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.38,
|
| 87 |
+
"learning_rate": 5e-05,
|
| 88 |
+
"loss": 2.8391,
|
| 89 |
+
"step": 6
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.38,
|
| 93 |
+
"eval_accuracy": 0.040284552845528454,
|
| 94 |
+
"eval_loss": 2.74609375,
|
| 95 |
+
"eval_runtime": 6.8467,
|
| 96 |
+
"eval_samples_per_second": 14.606,
|
| 97 |
+
"eval_steps_per_second": 0.292,
|
| 98 |
+
"step": 6
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.44,
|
| 102 |
+
"learning_rate": 5e-05,
|
| 103 |
+
"loss": 2.9368,
|
| 104 |
+
"step": 7
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.44,
|
| 108 |
+
"eval_accuracy": 0.03983739837398374,
|
| 109 |
+
"eval_loss": 2.720703125,
|
| 110 |
+
"eval_runtime": 7.2744,
|
| 111 |
+
"eval_samples_per_second": 13.747,
|
| 112 |
+
"eval_steps_per_second": 0.275,
|
| 113 |
+
"step": 7
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.5,
|
| 117 |
+
"learning_rate": 5e-05,
|
| 118 |
+
"loss": 2.7583,
|
| 119 |
+
"step": 8
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.5,
|
| 123 |
+
"eval_accuracy": 0.040325203252032524,
|
| 124 |
+
"eval_loss": 2.70703125,
|
| 125 |
+
"eval_runtime": 6.8783,
|
| 126 |
+
"eval_samples_per_second": 14.539,
|
| 127 |
+
"eval_steps_per_second": 0.291,
|
| 128 |
+
"step": 8
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.56,
|
| 132 |
+
"learning_rate": 5e-05,
|
| 133 |
+
"loss": 2.9756,
|
| 134 |
+
"step": 9
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.56,
|
| 138 |
+
"eval_accuracy": 0.04083333333333333,
|
| 139 |
+
"eval_loss": 2.68359375,
|
| 140 |
+
"eval_runtime": 7.0122,
|
| 141 |
+
"eval_samples_per_second": 14.261,
|
| 142 |
+
"eval_steps_per_second": 0.285,
|
| 143 |
+
"step": 9
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.62,
|
| 147 |
+
"learning_rate": 5e-05,
|
| 148 |
+
"loss": 2.8442,
|
| 149 |
+
"step": 10
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.62,
|
| 153 |
+
"eval_accuracy": 0.04034552845528455,
|
| 154 |
+
"eval_loss": 2.673828125,
|
| 155 |
+
"eval_runtime": 7.0273,
|
| 156 |
+
"eval_samples_per_second": 14.23,
|
| 157 |
+
"eval_steps_per_second": 0.285,
|
| 158 |
+
"step": 10
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.69,
|
| 162 |
+
"learning_rate": 5e-05,
|
| 163 |
+
"loss": 2.7312,
|
| 164 |
+
"step": 11
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.69,
|
| 168 |
+
"eval_accuracy": 0.04054878048780488,
|
| 169 |
+
"eval_loss": 2.66796875,
|
| 170 |
+
"eval_runtime": 6.8584,
|
| 171 |
+
"eval_samples_per_second": 14.581,
|
| 172 |
+
"eval_steps_per_second": 0.292,
|
| 173 |
+
"step": 11
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.75,
|
| 177 |
+
"learning_rate": 5e-05,
|
| 178 |
+
"loss": 2.7439,
|
| 179 |
+
"step": 12
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.75,
|
| 183 |
+
"eval_accuracy": 0.040365853658536587,
|
| 184 |
+
"eval_loss": 2.669921875,
|
| 185 |
+
"eval_runtime": 6.8135,
|
| 186 |
+
"eval_samples_per_second": 14.677,
|
| 187 |
+
"eval_steps_per_second": 0.294,
|
| 188 |
+
"step": 12
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.81,
|
| 192 |
+
"learning_rate": 5e-05,
|
| 193 |
+
"loss": 2.9075,
|
| 194 |
+
"step": 13
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.81,
|
| 198 |
+
"eval_accuracy": 0.04034552845528455,
|
| 199 |
+
"eval_loss": 2.6796875,
|
| 200 |
+
"eval_runtime": 6.1497,
|
| 201 |
+
"eval_samples_per_second": 16.261,
|
| 202 |
+
"eval_steps_per_second": 0.325,
|
| 203 |
+
"step": 13
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 0.88,
|
| 207 |
+
"learning_rate": 5e-05,
|
| 208 |
+
"loss": 2.8518,
|
| 209 |
+
"step": 14
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.88,
|
| 213 |
+
"eval_accuracy": 0.04034552845528455,
|
| 214 |
+
"eval_loss": 2.6796875,
|
| 215 |
+
"eval_runtime": 6.4271,
|
| 216 |
+
"eval_samples_per_second": 15.559,
|
| 217 |
+
"eval_steps_per_second": 0.311,
|
| 218 |
+
"step": 14
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.94,
|
| 222 |
+
"learning_rate": 5e-05,
|
| 223 |
+
"loss": 2.8579,
|
| 224 |
+
"step": 15
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.94,
|
| 228 |
+
"eval_accuracy": 0.04044715447154471,
|
| 229 |
+
"eval_loss": 2.677734375,
|
| 230 |
+
"eval_runtime": 7.2159,
|
| 231 |
+
"eval_samples_per_second": 13.858,
|
| 232 |
+
"eval_steps_per_second": 0.277,
|
| 233 |
+
"step": 15
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 1.0,
|
| 237 |
+
"learning_rate": 5e-05,
|
| 238 |
+
"loss": 2.8916,
|
| 239 |
+
"step": 16
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 1.0,
|
| 243 |
+
"eval_accuracy": 0.040325203252032524,
|
| 244 |
+
"eval_loss": 2.6953125,
|
| 245 |
+
"eval_runtime": 6.3862,
|
| 246 |
+
"eval_samples_per_second": 15.659,
|
| 247 |
+
"eval_steps_per_second": 0.313,
|
| 248 |
+
"step": 16
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.0,
|
| 252 |
+
"step": 16,
|
| 253 |
+
"total_flos": 4999961640960.0,
|
| 254 |
+
"train_loss": 2.8766326904296875,
|
| 255 |
+
"train_runtime": 267.0978,
|
| 256 |
+
"train_samples_per_second": 3.744,
|
| 257 |
+
"train_steps_per_second": 0.06
|
| 258 |
+
}
|
| 259 |
+
],
|
| 260 |
+
"max_steps": 16,
|
| 261 |
+
"num_train_epochs": 1,
|
| 262 |
+
"total_flos": 4999961640960.0,
|
| 263 |
+
"trial_name": null,
|
| 264 |
+
"trial_params": null
|
| 265 |
+
}
|