File size: 6,290 Bytes
b9f11f7 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 |
{
"best_metric": 0.8277886497064579,
"best_model_checkpoint": "tiny-bert-sst2-distilled/run-1/checkpoint-1920",
"epoch": 10.0,
"eval_steps": 500,
"global_step": 1920,
"is_hyper_param_search": true,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"grad_norm": 0.8947777152061462,
"learning_rate": 0.0005181908296523575,
"loss": 0.4457,
"step": 192
},
{
"epoch": 1.0,
"eval_accuracy": 0.7925636007827789,
"eval_f1": 0.8090090090090091,
"eval_loss": 0.41997030377388,
"eval_mcc": 0.59400157079309,
"eval_precision": 0.7495826377295493,
"eval_recall": 0.8786692759295499,
"eval_runtime": 66.1643,
"eval_samples_per_second": 15.446,
"eval_steps_per_second": 0.484,
"step": 192
},
{
"epoch": 2.0,
"grad_norm": 1.971238613128662,
"learning_rate": 0.0004606140708020956,
"loss": 0.4163,
"step": 384
},
{
"epoch": 2.0,
"eval_accuracy": 0.7915851272015656,
"eval_f1": 0.7957813998082454,
"eval_loss": 0.39958736300468445,
"eval_mcc": 0.5836633290639656,
"eval_precision": 0.7800751879699248,
"eval_recall": 0.812133072407045,
"eval_runtime": 66.5614,
"eval_samples_per_second": 15.354,
"eval_steps_per_second": 0.481,
"step": 384
},
{
"epoch": 3.0,
"grad_norm": 2.333700656890869,
"learning_rate": 0.00040303731195183364,
"loss": 0.4018,
"step": 576
},
{
"epoch": 3.0,
"eval_accuracy": 0.7915851272015656,
"eval_f1": 0.7909715407262021,
"eval_loss": 0.40145382285118103,
"eval_mcc": 0.5831803046632951,
"eval_precision": 0.7933070866141733,
"eval_recall": 0.7886497064579256,
"eval_runtime": 66.8237,
"eval_samples_per_second": 15.294,
"eval_steps_per_second": 0.479,
"step": 576
},
{
"epoch": 4.0,
"grad_norm": 1.9171998500823975,
"learning_rate": 0.00034546055310157167,
"loss": 0.3976,
"step": 768
},
{
"epoch": 4.0,
"eval_accuracy": 0.786692759295499,
"eval_f1": 0.783300198807157,
"eval_loss": 0.40033411979675293,
"eval_mcc": 0.5736667956847438,
"eval_precision": 0.795959595959596,
"eval_recall": 0.7710371819960861,
"eval_runtime": 66.6803,
"eval_samples_per_second": 15.327,
"eval_steps_per_second": 0.48,
"step": 768
},
{
"epoch": 5.0,
"grad_norm": 1.3183633089065552,
"learning_rate": 0.00028788379425130975,
"loss": 0.3869,
"step": 960
},
{
"epoch": 5.0,
"eval_accuracy": 0.8013698630136986,
"eval_f1": 0.8031037827352084,
"eval_loss": 0.39175575971603394,
"eval_mcc": 0.60283323302071,
"eval_precision": 0.7961538461538461,
"eval_recall": 0.8101761252446184,
"eval_runtime": 66.5306,
"eval_samples_per_second": 15.361,
"eval_steps_per_second": 0.481,
"step": 960
},
{
"epoch": 6.0,
"grad_norm": 1.770856499671936,
"learning_rate": 0.0002303070354010478,
"loss": 0.3843,
"step": 1152
},
{
"epoch": 6.0,
"eval_accuracy": 0.8228962818003914,
"eval_f1": 0.8350045578851412,
"eval_loss": 0.38628602027893066,
"eval_mcc": 0.6528627457969611,
"eval_precision": 0.7815699658703071,
"eval_recall": 0.8962818003913894,
"eval_runtime": 67.0562,
"eval_samples_per_second": 15.241,
"eval_steps_per_second": 0.477,
"step": 1152
},
{
"epoch": 7.0,
"grad_norm": 1.1659635305404663,
"learning_rate": 0.00017273027655078583,
"loss": 0.3793,
"step": 1344
},
{
"epoch": 7.0,
"eval_accuracy": 0.8111545988258317,
"eval_f1": 0.8269058295964126,
"eval_loss": 0.39157727360725403,
"eval_mcc": 0.6328787442530197,
"eval_precision": 0.7632450331125827,
"eval_recall": 0.9021526418786693,
"eval_runtime": 66.5334,
"eval_samples_per_second": 15.361,
"eval_steps_per_second": 0.481,
"step": 1344
},
{
"epoch": 8.0,
"grad_norm": 2.5721819400787354,
"learning_rate": 0.0001151535177005239,
"loss": 0.3727,
"step": 1536
},
{
"epoch": 8.0,
"eval_accuracy": 0.821917808219178,
"eval_f1": 0.8311688311688312,
"eval_loss": 0.3822523355484009,
"eval_mcc": 0.6477369472991997,
"eval_precision": 0.7901234567901234,
"eval_recall": 0.8767123287671232,
"eval_runtime": 67.4301,
"eval_samples_per_second": 15.156,
"eval_steps_per_second": 0.475,
"step": 1536
},
{
"epoch": 9.0,
"grad_norm": 1.5437999963760376,
"learning_rate": 5.757675885026195e-05,
"loss": 0.3692,
"step": 1728
},
{
"epoch": 9.0,
"eval_accuracy": 0.8268101761252447,
"eval_f1": 0.8334901222953904,
"eval_loss": 0.37901678681373596,
"eval_mcc": 0.655734447578894,
"eval_precision": 0.802536231884058,
"eval_recall": 0.8669275929549902,
"eval_runtime": 67.6441,
"eval_samples_per_second": 15.108,
"eval_steps_per_second": 0.473,
"step": 1728
},
{
"epoch": 10.0,
"grad_norm": 1.2340621948242188,
"learning_rate": 0.0,
"loss": 0.3673,
"step": 1920
},
{
"epoch": 10.0,
"eval_accuracy": 0.8277886497064579,
"eval_f1": 0.8352059925093633,
"eval_loss": 0.37935787439346313,
"eval_mcc": 0.6582497957275898,
"eval_precision": 0.800718132854578,
"eval_recall": 0.87279843444227,
"eval_runtime": 67.2953,
"eval_samples_per_second": 15.187,
"eval_steps_per_second": 0.476,
"step": 1920
}
],
"logging_steps": 500,
"max_steps": 1920,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 500,
"total_flos": 2356951972800.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": {
"alpha": 0.6664545915670365,
"learning_rate": 0.0005757675885026195,
"num_train_epochs": 10,
"per_device_train_batch_size": 16,
"temperature": 37
}
}
|