Commit
•
43fb6a6
1
Parent(s):
a1fa2c6
Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +262 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,262 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.08392839133739471,
|
| 3 |
+
"best_model_checkpoint": "./models/results_funlines_and_human_microedit_paper_2023_693/checkpoint-200",
|
| 4 |
+
"epoch": 0.774818401937046,
|
| 5 |
+
"global_step": 200,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.03,
|
| 12 |
+
"learning_rate": 4.000000000000001e-06,
|
| 13 |
+
"loss": 0.6859,
|
| 14 |
+
"step": 8
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.06,
|
| 18 |
+
"learning_rate": 8.000000000000001e-06,
|
| 19 |
+
"loss": 0.6827,
|
| 20 |
+
"step": 16
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.09,
|
| 24 |
+
"learning_rate": 1.2e-05,
|
| 25 |
+
"loss": 0.6877,
|
| 26 |
+
"step": 24
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.1,
|
| 30 |
+
"eval_accuracy": 0.5563559322033899,
|
| 31 |
+
"eval_f1": 0.7149469098829294,
|
| 32 |
+
"eval_loss": 0.6760356426239014,
|
| 33 |
+
"eval_precision": 0.5563559322033899,
|
| 34 |
+
"eval_recall": 1.0,
|
| 35 |
+
"eval_runtime": 15.2381,
|
| 36 |
+
"eval_samples_per_second": 154.875,
|
| 37 |
+
"eval_steps_per_second": 19.359,
|
| 38 |
+
"step": 25
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.12,
|
| 42 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 43 |
+
"loss": 0.6647,
|
| 44 |
+
"step": 32
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.15,
|
| 48 |
+
"learning_rate": 2e-05,
|
| 49 |
+
"loss": 0.543,
|
| 50 |
+
"step": 40
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.19,
|
| 54 |
+
"learning_rate": 2.4e-05,
|
| 55 |
+
"loss": 0.3736,
|
| 56 |
+
"step": 48
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.19,
|
| 60 |
+
"eval_accuracy": 0.8635593220338983,
|
| 61 |
+
"eval_f1": 0.8649328859060402,
|
| 62 |
+
"eval_loss": 0.31369638442993164,
|
| 63 |
+
"eval_precision": 0.9626517273576097,
|
| 64 |
+
"eval_recall": 0.7852246763137852,
|
| 65 |
+
"eval_runtime": 15.1929,
|
| 66 |
+
"eval_samples_per_second": 155.336,
|
| 67 |
+
"eval_steps_per_second": 19.417,
|
| 68 |
+
"step": 50
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.22,
|
| 72 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 73 |
+
"loss": 0.3738,
|
| 74 |
+
"step": 56
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.25,
|
| 78 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 79 |
+
"loss": 0.2697,
|
| 80 |
+
"step": 64
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.28,
|
| 84 |
+
"learning_rate": 3.6e-05,
|
| 85 |
+
"loss": 0.1821,
|
| 86 |
+
"step": 72
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.29,
|
| 90 |
+
"eval_accuracy": 0.9555084745762712,
|
| 91 |
+
"eval_f1": 0.9600608596424496,
|
| 92 |
+
"eval_loss": 0.13926365971565247,
|
| 93 |
+
"eval_precision": 0.958966565349544,
|
| 94 |
+
"eval_recall": 0.9611576542269612,
|
| 95 |
+
"eval_runtime": 15.1835,
|
| 96 |
+
"eval_samples_per_second": 155.432,
|
| 97 |
+
"eval_steps_per_second": 19.429,
|
| 98 |
+
"step": 75
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.31,
|
| 102 |
+
"learning_rate": 3.9000000000000006e-05,
|
| 103 |
+
"loss": 0.1583,
|
| 104 |
+
"step": 80
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.34,
|
| 108 |
+
"learning_rate": 4.3e-05,
|
| 109 |
+
"loss": 0.1845,
|
| 110 |
+
"step": 88
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.37,
|
| 114 |
+
"learning_rate": 4.7e-05,
|
| 115 |
+
"loss": 0.1605,
|
| 116 |
+
"step": 96
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.39,
|
| 120 |
+
"eval_accuracy": 0.9593220338983051,
|
| 121 |
+
"eval_f1": 0.9625,
|
| 122 |
+
"eval_loss": 0.13289286196231842,
|
| 123 |
+
"eval_precision": 0.9879711307137129,
|
| 124 |
+
"eval_recall": 0.9383092155369384,
|
| 125 |
+
"eval_runtime": 15.1752,
|
| 126 |
+
"eval_samples_per_second": 155.517,
|
| 127 |
+
"eval_steps_per_second": 19.44,
|
| 128 |
+
"step": 100
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.4,
|
| 132 |
+
"learning_rate": 4.936708860759494e-05,
|
| 133 |
+
"loss": 0.1174,
|
| 134 |
+
"step": 104
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.43,
|
| 138 |
+
"learning_rate": 4.683544303797468e-05,
|
| 139 |
+
"loss": 0.1111,
|
| 140 |
+
"step": 112
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.46,
|
| 144 |
+
"learning_rate": 4.430379746835443e-05,
|
| 145 |
+
"loss": 0.158,
|
| 146 |
+
"step": 120
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.48,
|
| 150 |
+
"eval_accuracy": 0.9567796610169491,
|
| 151 |
+
"eval_f1": 0.9617977528089888,
|
| 152 |
+
"eval_loss": 0.11806055903434753,
|
| 153 |
+
"eval_precision": 0.94620486366986,
|
| 154 |
+
"eval_recall": 0.9779131759329779,
|
| 155 |
+
"eval_runtime": 15.1814,
|
| 156 |
+
"eval_samples_per_second": 155.454,
|
| 157 |
+
"eval_steps_per_second": 19.432,
|
| 158 |
+
"step": 125
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.5,
|
| 162 |
+
"learning_rate": 4.177215189873418e-05,
|
| 163 |
+
"loss": 0.1373,
|
| 164 |
+
"step": 128
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.53,
|
| 168 |
+
"learning_rate": 3.924050632911392e-05,
|
| 169 |
+
"loss": 0.1306,
|
| 170 |
+
"step": 136
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.56,
|
| 174 |
+
"learning_rate": 3.670886075949367e-05,
|
| 175 |
+
"loss": 0.1261,
|
| 176 |
+
"step": 144
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.58,
|
| 180 |
+
"eval_accuracy": 0.9669491525423729,
|
| 181 |
+
"eval_f1": 0.9698841698841698,
|
| 182 |
+
"eval_loss": 0.09348666667938232,
|
| 183 |
+
"eval_precision": 0.9835552075176194,
|
| 184 |
+
"eval_recall": 0.9565879664889566,
|
| 185 |
+
"eval_runtime": 15.1738,
|
| 186 |
+
"eval_samples_per_second": 155.531,
|
| 187 |
+
"eval_steps_per_second": 19.441,
|
| 188 |
+
"step": 150
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.59,
|
| 192 |
+
"learning_rate": 3.4177215189873416e-05,
|
| 193 |
+
"loss": 0.1048,
|
| 194 |
+
"step": 152
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.62,
|
| 198 |
+
"learning_rate": 3.1645569620253167e-05,
|
| 199 |
+
"loss": 0.1109,
|
| 200 |
+
"step": 160
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.65,
|
| 204 |
+
"learning_rate": 2.9113924050632914e-05,
|
| 205 |
+
"loss": 0.0743,
|
| 206 |
+
"step": 168
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.68,
|
| 210 |
+
"eval_accuracy": 0.9703389830508474,
|
| 211 |
+
"eval_f1": 0.9728682170542636,
|
| 212 |
+
"eval_loss": 0.10754456371068954,
|
| 213 |
+
"eval_precision": 0.9905288082083662,
|
| 214 |
+
"eval_recall": 0.9558263518659559,
|
| 215 |
+
"eval_runtime": 15.1659,
|
| 216 |
+
"eval_samples_per_second": 155.612,
|
| 217 |
+
"eval_steps_per_second": 19.451,
|
| 218 |
+
"step": 175
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.68,
|
| 222 |
+
"learning_rate": 2.6582278481012658e-05,
|
| 223 |
+
"loss": 0.1677,
|
| 224 |
+
"step": 176
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.71,
|
| 228 |
+
"learning_rate": 2.4050632911392405e-05,
|
| 229 |
+
"loss": 0.0963,
|
| 230 |
+
"step": 184
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.74,
|
| 234 |
+
"learning_rate": 2.1518987341772153e-05,
|
| 235 |
+
"loss": 0.0954,
|
| 236 |
+
"step": 192
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.77,
|
| 240 |
+
"learning_rate": 1.89873417721519e-05,
|
| 241 |
+
"loss": 0.1263,
|
| 242 |
+
"step": 200
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.77,
|
| 246 |
+
"eval_accuracy": 0.9699152542372881,
|
| 247 |
+
"eval_f1": 0.9728904161893853,
|
| 248 |
+
"eval_loss": 0.08392839133739471,
|
| 249 |
+
"eval_precision": 0.9754977029096478,
|
| 250 |
+
"eval_recall": 0.9702970297029703,
|
| 251 |
+
"eval_runtime": 15.2112,
|
| 252 |
+
"eval_samples_per_second": 155.148,
|
| 253 |
+
"eval_steps_per_second": 19.394,
|
| 254 |
+
"step": 200
|
| 255 |
+
}
|
| 256 |
+
],
|
| 257 |
+
"max_steps": 258,
|
| 258 |
+
"num_train_epochs": 1,
|
| 259 |
+
"total_flos": 289422160896000.0,
|
| 260 |
+
"trial_name": null,
|
| 261 |
+
"trial_params": null
|
| 262 |
+
}
|