Commit
•
f12c3da
1
Parent(s):
085f356
Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +277 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,277 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 20.0,
|
| 5 |
+
"global_step": 6220,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.0,
|
| 12 |
+
"eval_accuracy": 0.9522821307182312,
|
| 13 |
+
"eval_loss": 0.12891167402267456,
|
| 14 |
+
"eval_runtime": 9.8242,
|
| 15 |
+
"eval_samples_per_second": 147.188,
|
| 16 |
+
"eval_steps_per_second": 18.424,
|
| 17 |
+
"step": 311
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 1.61,
|
| 21 |
+
"learning_rate": 3.106972231499722e-05,
|
| 22 |
+
"loss": 0.2799,
|
| 23 |
+
"step": 500
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 2.0,
|
| 27 |
+
"eval_accuracy": 0.9612724781036377,
|
| 28 |
+
"eval_loss": 0.11741334199905396,
|
| 29 |
+
"eval_runtime": 9.9246,
|
| 30 |
+
"eval_samples_per_second": 145.698,
|
| 31 |
+
"eval_steps_per_second": 18.237,
|
| 32 |
+
"step": 622
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 3.0,
|
| 36 |
+
"eval_accuracy": 0.9688796401023865,
|
| 37 |
+
"eval_loss": 0.1398700326681137,
|
| 38 |
+
"eval_runtime": 9.777,
|
| 39 |
+
"eval_samples_per_second": 147.898,
|
| 40 |
+
"eval_steps_per_second": 18.513,
|
| 41 |
+
"step": 933
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 3.22,
|
| 45 |
+
"learning_rate": 2.8353837497252705e-05,
|
| 46 |
+
"loss": 0.0986,
|
| 47 |
+
"step": 1000
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 4.0,
|
| 51 |
+
"eval_accuracy": 0.9640387296676636,
|
| 52 |
+
"eval_loss": 0.19078733026981354,
|
| 53 |
+
"eval_runtime": 10.086,
|
| 54 |
+
"eval_samples_per_second": 143.368,
|
| 55 |
+
"eval_steps_per_second": 17.946,
|
| 56 |
+
"step": 1244
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 4.82,
|
| 60 |
+
"learning_rate": 2.563795267950819e-05,
|
| 61 |
+
"loss": 0.0467,
|
| 62 |
+
"step": 1500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 5.0,
|
| 66 |
+
"eval_accuracy": 0.9598893523216248,
|
| 67 |
+
"eval_loss": 0.2066243439912796,
|
| 68 |
+
"eval_runtime": 9.9772,
|
| 69 |
+
"eval_samples_per_second": 144.93,
|
| 70 |
+
"eval_steps_per_second": 18.141,
|
| 71 |
+
"step": 1555
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 6.0,
|
| 75 |
+
"eval_accuracy": 0.9619640111923218,
|
| 76 |
+
"eval_loss": 0.23871752619743347,
|
| 77 |
+
"eval_runtime": 9.7438,
|
| 78 |
+
"eval_samples_per_second": 148.402,
|
| 79 |
+
"eval_steps_per_second": 18.576,
|
| 80 |
+
"step": 1866
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 6.43,
|
| 84 |
+
"learning_rate": 2.292206786176368e-05,
|
| 85 |
+
"loss": 0.0299,
|
| 86 |
+
"step": 2000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 7.0,
|
| 90 |
+
"eval_accuracy": 0.9640387296676636,
|
| 91 |
+
"eval_loss": 0.2593800723552704,
|
| 92 |
+
"eval_runtime": 9.8624,
|
| 93 |
+
"eval_samples_per_second": 146.617,
|
| 94 |
+
"eval_steps_per_second": 18.352,
|
| 95 |
+
"step": 2177
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 8.0,
|
| 99 |
+
"eval_accuracy": 0.9681881070137024,
|
| 100 |
+
"eval_loss": 0.21497485041618347,
|
| 101 |
+
"eval_runtime": 9.6756,
|
| 102 |
+
"eval_samples_per_second": 149.448,
|
| 103 |
+
"eval_steps_per_second": 18.707,
|
| 104 |
+
"step": 2488
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 8.04,
|
| 108 |
+
"learning_rate": 2.020618304401917e-05,
|
| 109 |
+
"loss": 0.0174,
|
| 110 |
+
"step": 2500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 9.0,
|
| 114 |
+
"eval_accuracy": 0.9702627658843994,
|
| 115 |
+
"eval_loss": 0.21150082349777222,
|
| 116 |
+
"eval_runtime": 10.0341,
|
| 117 |
+
"eval_samples_per_second": 144.108,
|
| 118 |
+
"eval_steps_per_second": 18.038,
|
| 119 |
+
"step": 2799
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 9.65,
|
| 123 |
+
"learning_rate": 1.749029822627466e-05,
|
| 124 |
+
"loss": 0.0133,
|
| 125 |
+
"step": 3000
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 10.0,
|
| 129 |
+
"eval_accuracy": 0.9661133885383606,
|
| 130 |
+
"eval_loss": 0.2336471825838089,
|
| 131 |
+
"eval_runtime": 9.6376,
|
| 132 |
+
"eval_samples_per_second": 150.037,
|
| 133 |
+
"eval_steps_per_second": 18.781,
|
| 134 |
+
"step": 3110
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 11.0,
|
| 138 |
+
"eval_accuracy": 0.9640387296676636,
|
| 139 |
+
"eval_loss": 0.26857590675354004,
|
| 140 |
+
"eval_runtime": 10.0374,
|
| 141 |
+
"eval_samples_per_second": 144.061,
|
| 142 |
+
"eval_steps_per_second": 18.033,
|
| 143 |
+
"step": 3421
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 11.25,
|
| 147 |
+
"learning_rate": 1.4774413408530146e-05,
|
| 148 |
+
"loss": 0.0065,
|
| 149 |
+
"step": 3500
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 12.0,
|
| 153 |
+
"eval_accuracy": 0.9647302627563477,
|
| 154 |
+
"eval_loss": 0.29211708903312683,
|
| 155 |
+
"eval_runtime": 9.8275,
|
| 156 |
+
"eval_samples_per_second": 147.138,
|
| 157 |
+
"eval_steps_per_second": 18.418,
|
| 158 |
+
"step": 3732
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 12.86,
|
| 162 |
+
"learning_rate": 1.2058528590785633e-05,
|
| 163 |
+
"loss": 0.006,
|
| 164 |
+
"step": 4000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 13.0,
|
| 168 |
+
"eval_accuracy": 0.9633471369743347,
|
| 169 |
+
"eval_loss": 0.30721694231033325,
|
| 170 |
+
"eval_runtime": 9.9209,
|
| 171 |
+
"eval_samples_per_second": 145.753,
|
| 172 |
+
"eval_steps_per_second": 18.244,
|
| 173 |
+
"step": 4043
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 14.0,
|
| 177 |
+
"eval_accuracy": 0.9674965143203735,
|
| 178 |
+
"eval_loss": 0.30118855834007263,
|
| 179 |
+
"eval_runtime": 9.8913,
|
| 180 |
+
"eval_samples_per_second": 146.189,
|
| 181 |
+
"eval_steps_per_second": 18.299,
|
| 182 |
+
"step": 4354
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 14.47,
|
| 186 |
+
"learning_rate": 9.342643773041121e-06,
|
| 187 |
+
"loss": 0.0068,
|
| 188 |
+
"step": 4500
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 15.0,
|
| 192 |
+
"eval_accuracy": 0.9647302627563477,
|
| 193 |
+
"eval_loss": 0.29607123136520386,
|
| 194 |
+
"eval_runtime": 9.8325,
|
| 195 |
+
"eval_samples_per_second": 147.063,
|
| 196 |
+
"eval_steps_per_second": 18.408,
|
| 197 |
+
"step": 4665
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 16.0,
|
| 201 |
+
"eval_accuracy": 0.9633471369743347,
|
| 202 |
+
"eval_loss": 0.29950061440467834,
|
| 203 |
+
"eval_runtime": 9.895,
|
| 204 |
+
"eval_samples_per_second": 146.134,
|
| 205 |
+
"eval_steps_per_second": 18.292,
|
| 206 |
+
"step": 4976
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 16.08,
|
| 210 |
+
"learning_rate": 6.626758955296609e-06,
|
| 211 |
+
"loss": 0.0062,
|
| 212 |
+
"step": 5000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 17.0,
|
| 216 |
+
"eval_accuracy": 0.9674965143203735,
|
| 217 |
+
"eval_loss": 0.2802109718322754,
|
| 218 |
+
"eval_runtime": 9.7901,
|
| 219 |
+
"eval_samples_per_second": 147.7,
|
| 220 |
+
"eval_steps_per_second": 18.488,
|
| 221 |
+
"step": 5287
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 17.68,
|
| 225 |
+
"learning_rate": 3.910874137552097e-06,
|
| 226 |
+
"loss": 0.0069,
|
| 227 |
+
"step": 5500
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 18.0,
|
| 231 |
+
"eval_accuracy": 0.9688796401023865,
|
| 232 |
+
"eval_loss": 0.2822701930999756,
|
| 233 |
+
"eval_runtime": 9.8995,
|
| 234 |
+
"eval_samples_per_second": 146.068,
|
| 235 |
+
"eval_steps_per_second": 18.284,
|
| 236 |
+
"step": 5598
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 19.0,
|
| 240 |
+
"eval_accuracy": 0.9681881070137024,
|
| 241 |
+
"eval_loss": 0.27318963408470154,
|
| 242 |
+
"eval_runtime": 9.6783,
|
| 243 |
+
"eval_samples_per_second": 149.407,
|
| 244 |
+
"eval_steps_per_second": 18.702,
|
| 245 |
+
"step": 5909
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 19.29,
|
| 249 |
+
"learning_rate": 1.1949893198075852e-06,
|
| 250 |
+
"loss": 0.0041,
|
| 251 |
+
"step": 6000
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 20.0,
|
| 255 |
+
"eval_accuracy": 0.9681881070137024,
|
| 256 |
+
"eval_loss": 0.27601131796836853,
|
| 257 |
+
"eval_runtime": 9.9846,
|
| 258 |
+
"eval_samples_per_second": 144.824,
|
| 259 |
+
"eval_steps_per_second": 18.128,
|
| 260 |
+
"step": 6220
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 20.0,
|
| 264 |
+
"step": 6220,
|
| 265 |
+
"total_flos": 6546379231641600.0,
|
| 266 |
+
"train_loss": 0.042026480821551235,
|
| 267 |
+
"train_runtime": 4438.5906,
|
| 268 |
+
"train_samples_per_second": 22.422,
|
| 269 |
+
"train_steps_per_second": 1.401
|
| 270 |
+
}
|
| 271 |
+
],
|
| 272 |
+
"max_steps": 6220,
|
| 273 |
+
"num_train_epochs": 20,
|
| 274 |
+
"total_flos": 6546379231641600.0,
|
| 275 |
+
"trial_name": null,
|
| 276 |
+
"trial_params": null
|
| 277 |
+
}
|