Commit
•
f52b213
1
Parent(s):
221700a
Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +271 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,271 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 20.0,
|
| 5 |
+
"global_step": 5600,
|
| 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": 0.9008928537368774,
|
| 13 |
+
"eval_loss": 0.2425430864095688,
|
| 14 |
+
"eval_runtime": 15.3273,
|
| 15 |
+
"eval_samples_per_second": 73.072,
|
| 16 |
+
"eval_steps_per_second": 9.134,
|
| 17 |
+
"step": 280
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 1.79,
|
| 21 |
+
"learning_rate": 3.0769035067318356e-05,
|
| 22 |
+
"loss": 0.2497,
|
| 23 |
+
"step": 500
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 2.0,
|
| 27 |
+
"eval_accuracy": 0.90625,
|
| 28 |
+
"eval_loss": 0.2671281397342682,
|
| 29 |
+
"eval_runtime": 15.4851,
|
| 30 |
+
"eval_samples_per_second": 72.327,
|
| 31 |
+
"eval_steps_per_second": 9.041,
|
| 32 |
+
"step": 560
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 3.0,
|
| 36 |
+
"eval_accuracy": 0.9285714030265808,
|
| 37 |
+
"eval_loss": 0.37120091915130615,
|
| 38 |
+
"eval_runtime": 15.3992,
|
| 39 |
+
"eval_samples_per_second": 72.731,
|
| 40 |
+
"eval_steps_per_second": 9.091,
|
| 41 |
+
"step": 840
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 3.57,
|
| 45 |
+
"learning_rate": 2.775246300189499e-05,
|
| 46 |
+
"loss": 0.0886,
|
| 47 |
+
"step": 1000
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 4.0,
|
| 51 |
+
"eval_accuracy": 0.9357143044471741,
|
| 52 |
+
"eval_loss": 0.3282666504383087,
|
| 53 |
+
"eval_runtime": 15.4636,
|
| 54 |
+
"eval_samples_per_second": 72.428,
|
| 55 |
+
"eval_steps_per_second": 9.054,
|
| 56 |
+
"step": 1120
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 5.0,
|
| 60 |
+
"eval_accuracy": 0.9303571581840515,
|
| 61 |
+
"eval_loss": 0.36973103880882263,
|
| 62 |
+
"eval_runtime": 15.65,
|
| 63 |
+
"eval_samples_per_second": 71.565,
|
| 64 |
+
"eval_steps_per_second": 8.946,
|
| 65 |
+
"step": 1400
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 5.36,
|
| 69 |
+
"learning_rate": 2.473589093647162e-05,
|
| 70 |
+
"loss": 0.0277,
|
| 71 |
+
"step": 1500
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 6.0,
|
| 75 |
+
"eval_accuracy": 0.9303571581840515,
|
| 76 |
+
"eval_loss": 0.4028499722480774,
|
| 77 |
+
"eval_runtime": 15.6492,
|
| 78 |
+
"eval_samples_per_second": 71.569,
|
| 79 |
+
"eval_steps_per_second": 8.946,
|
| 80 |
+
"step": 1680
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 7.0,
|
| 84 |
+
"eval_accuracy": 0.9285714030265808,
|
| 85 |
+
"eval_loss": 0.4771631956100464,
|
| 86 |
+
"eval_runtime": 15.5073,
|
| 87 |
+
"eval_samples_per_second": 72.224,
|
| 88 |
+
"eval_steps_per_second": 9.028,
|
| 89 |
+
"step": 1960
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 7.14,
|
| 93 |
+
"learning_rate": 2.1719318871048256e-05,
|
| 94 |
+
"loss": 0.023,
|
| 95 |
+
"step": 2000
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 8.0,
|
| 99 |
+
"eval_accuracy": 0.9205357432365417,
|
| 100 |
+
"eval_loss": 0.5438315868377686,
|
| 101 |
+
"eval_runtime": 15.6912,
|
| 102 |
+
"eval_samples_per_second": 71.378,
|
| 103 |
+
"eval_steps_per_second": 8.922,
|
| 104 |
+
"step": 2240
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 8.93,
|
| 108 |
+
"learning_rate": 1.8702746805624887e-05,
|
| 109 |
+
"loss": 0.0229,
|
| 110 |
+
"step": 2500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 9.0,
|
| 114 |
+
"eval_accuracy": 0.9276785850524902,
|
| 115 |
+
"eval_loss": 0.51451176404953,
|
| 116 |
+
"eval_runtime": 15.4278,
|
| 117 |
+
"eval_samples_per_second": 72.596,
|
| 118 |
+
"eval_steps_per_second": 9.075,
|
| 119 |
+
"step": 2520
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 10.0,
|
| 123 |
+
"eval_accuracy": 0.9223214387893677,
|
| 124 |
+
"eval_loss": 0.4909174144268036,
|
| 125 |
+
"eval_runtime": 15.4206,
|
| 126 |
+
"eval_samples_per_second": 72.63,
|
| 127 |
+
"eval_steps_per_second": 9.079,
|
| 128 |
+
"step": 2800
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 10.71,
|
| 132 |
+
"learning_rate": 1.5686174740201518e-05,
|
| 133 |
+
"loss": 0.0151,
|
| 134 |
+
"step": 3000
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 11.0,
|
| 138 |
+
"eval_accuracy": 0.9205357432365417,
|
| 139 |
+
"eval_loss": 0.5145156979560852,
|
| 140 |
+
"eval_runtime": 15.4916,
|
| 141 |
+
"eval_samples_per_second": 72.297,
|
| 142 |
+
"eval_steps_per_second": 9.037,
|
| 143 |
+
"step": 3080
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 12.0,
|
| 147 |
+
"eval_accuracy": 0.9241071343421936,
|
| 148 |
+
"eval_loss": 0.5320212244987488,
|
| 149 |
+
"eval_runtime": 15.5222,
|
| 150 |
+
"eval_samples_per_second": 72.155,
|
| 151 |
+
"eval_steps_per_second": 9.019,
|
| 152 |
+
"step": 3360
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 12.5,
|
| 156 |
+
"learning_rate": 1.2669602674778149e-05,
|
| 157 |
+
"loss": 0.0164,
|
| 158 |
+
"step": 3500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 13.0,
|
| 162 |
+
"eval_accuracy": 0.925000011920929,
|
| 163 |
+
"eval_loss": 0.4845915138721466,
|
| 164 |
+
"eval_runtime": 15.5092,
|
| 165 |
+
"eval_samples_per_second": 72.215,
|
| 166 |
+
"eval_steps_per_second": 9.027,
|
| 167 |
+
"step": 3640
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 14.0,
|
| 171 |
+
"eval_accuracy": 0.9232142567634583,
|
| 172 |
+
"eval_loss": 0.5342178344726562,
|
| 173 |
+
"eval_runtime": 15.4696,
|
| 174 |
+
"eval_samples_per_second": 72.4,
|
| 175 |
+
"eval_steps_per_second": 9.05,
|
| 176 |
+
"step": 3920
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 14.29,
|
| 180 |
+
"learning_rate": 9.653030609354778e-06,
|
| 181 |
+
"loss": 0.0114,
|
| 182 |
+
"step": 4000
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 15.0,
|
| 186 |
+
"eval_accuracy": 0.9241071343421936,
|
| 187 |
+
"eval_loss": 0.5672621726989746,
|
| 188 |
+
"eval_runtime": 15.5099,
|
| 189 |
+
"eval_samples_per_second": 72.212,
|
| 190 |
+
"eval_steps_per_second": 9.027,
|
| 191 |
+
"step": 4200
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 16.0,
|
| 195 |
+
"eval_accuracy": 0.9258928298950195,
|
| 196 |
+
"eval_loss": 0.562305748462677,
|
| 197 |
+
"eval_runtime": 15.6284,
|
| 198 |
+
"eval_samples_per_second": 71.664,
|
| 199 |
+
"eval_steps_per_second": 8.958,
|
| 200 |
+
"step": 4480
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 16.07,
|
| 204 |
+
"learning_rate": 6.636458543931411e-06,
|
| 205 |
+
"loss": 0.0087,
|
| 206 |
+
"step": 4500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 17.0,
|
| 210 |
+
"eval_accuracy": 0.9258928298950195,
|
| 211 |
+
"eval_loss": 0.5690274238586426,
|
| 212 |
+
"eval_runtime": 15.6003,
|
| 213 |
+
"eval_samples_per_second": 71.793,
|
| 214 |
+
"eval_steps_per_second": 8.974,
|
| 215 |
+
"step": 4760
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 17.86,
|
| 219 |
+
"learning_rate": 3.619886478508042e-06,
|
| 220 |
+
"loss": 0.0086,
|
| 221 |
+
"step": 5000
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 18.0,
|
| 225 |
+
"eval_accuracy": 0.925000011920929,
|
| 226 |
+
"eval_loss": 0.5821966528892517,
|
| 227 |
+
"eval_runtime": 15.4277,
|
| 228 |
+
"eval_samples_per_second": 72.597,
|
| 229 |
+
"eval_steps_per_second": 9.075,
|
| 230 |
+
"step": 5040
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 19.0,
|
| 234 |
+
"eval_accuracy": 0.925000011920929,
|
| 235 |
+
"eval_loss": 0.5867368578910828,
|
| 236 |
+
"eval_runtime": 15.4548,
|
| 237 |
+
"eval_samples_per_second": 72.469,
|
| 238 |
+
"eval_steps_per_second": 9.059,
|
| 239 |
+
"step": 5320
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 19.64,
|
| 243 |
+
"learning_rate": 6.033144130846736e-07,
|
| 244 |
+
"loss": 0.0067,
|
| 245 |
+
"step": 5500
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 20.0,
|
| 249 |
+
"eval_accuracy": 0.9258928298950195,
|
| 250 |
+
"eval_loss": 0.5937979221343994,
|
| 251 |
+
"eval_runtime": 15.5033,
|
| 252 |
+
"eval_samples_per_second": 72.243,
|
| 253 |
+
"eval_steps_per_second": 9.03,
|
| 254 |
+
"step": 5600
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 20.0,
|
| 258 |
+
"step": 5600,
|
| 259 |
+
"total_flos": 2.08752625385472e+16,
|
| 260 |
+
"train_loss": 0.042873825313789506,
|
| 261 |
+
"train_runtime": 6011.0282,
|
| 262 |
+
"train_samples_per_second": 14.906,
|
| 263 |
+
"train_steps_per_second": 0.932
|
| 264 |
+
}
|
| 265 |
+
],
|
| 266 |
+
"max_steps": 5600,
|
| 267 |
+
"num_train_epochs": 20,
|
| 268 |
+
"total_flos": 2.08752625385472e+16,
|
| 269 |
+
"trial_name": null,
|
| 270 |
+
"trial_params": null
|
| 271 |
+
}
|