| { | |
| "best_metric": 1.6784825325012207, | |
| "best_model_checkpoint": "output/headie-one/checkpoint-423", | |
| "epoch": 9.0, | |
| "global_step": 423, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00013340429968430322, | |
| "loss": 3.3897, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.00012243723783011615, | |
| "loss": 3.2804, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00010551244942700737, | |
| "loss": 3.1867, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 8.450286230835245e-05, | |
| "loss": 3.1078, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 6.173343555458685e-05, | |
| "loss": 2.9997, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 3.972387529741623e-05, | |
| "loss": 3.037, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 2.090979946151954e-05, | |
| "loss": 2.8067, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 7.373207860012155e-06, | |
| "loss": 2.6461, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 6.1208431258209e-07, | |
| "loss": 2.7311, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.646137237548828, | |
| "eval_runtime": 1.4447, | |
| "eval_samples_per_second": 46.378, | |
| "eval_steps_per_second": 6.23, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 1.3746270344901413e-06, | |
| "loss": 2.8414, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 9.576451662754438e-06, | |
| "loss": 2.7047, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 2.4309929383066146e-05, | |
| "loss": 2.7129, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 4.3944626783346644e-05, | |
| "loss": 2.7639, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 6.630773257727353e-05, | |
| "loss": 2.67, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 8.892450484875447e-05, | |
| "loss": 2.4349, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 0.00010929213048843373, | |
| "loss": 2.6474, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 0.00012515669103944476, | |
| "loss": 2.741, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 0.00013476258540873022, | |
| "loss": 2.5921, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.3871500492095947, | |
| "eval_runtime": 1.4491, | |
| "eval_samples_per_second": 46.235, | |
| "eval_steps_per_second": 6.211, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 0.00013704680787354832, | |
| "loss": 2.4265, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 0.00013175658222600302, | |
| "loss": 2.5165, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 0.00011947733444744994, | |
| "loss": 2.1946, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 0.0001015679084058065, | |
| "loss": 2.3182, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 8.001019372440279e-05, | |
| "loss": 2.2781, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 5.718980627559731e-05, | |
| "loss": 2.2824, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 3.563209159419354e-05, | |
| "loss": 2.2139, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 1.772266555255008e-05, | |
| "loss": 2.3554, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 5.443417773996978e-06, | |
| "loss": 2.1634, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 1.5319212645169297e-07, | |
| "loss": 2.2246, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.2159550189971924, | |
| "eval_runtime": 1.505, | |
| "eval_samples_per_second": 44.519, | |
| "eval_steps_per_second": 5.98, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 2.4374145912697595e-06, | |
| "loss": 2.0033, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 1.204330896055522e-05, | |
| "loss": 2.1305, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 2.790786951156628e-05, | |
| "loss": 2.0744, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 4.8275495151245426e-05, | |
| "loss": 2.1473, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 7.089226742272638e-05, | |
| "loss": 2.2506, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 9.325537321665337e-05, | |
| "loss": 2.157, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 0.00011289007061693382, | |
| "loss": 2.1313, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 0.00012762354833724553, | |
| "loss": 2.0593, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 0.00013582537296550986, | |
| "loss": 2.0425, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.1005759239196777, | |
| "eval_runtime": 1.5665, | |
| "eval_samples_per_second": 42.771, | |
| "eval_steps_per_second": 5.745, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 0.0001365879156874179, | |
| "loss": 1.9915, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 0.00012982679213998792, | |
| "loss": 1.8713, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 0.00011629020053848047, | |
| "loss": 1.8176, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 9.747612470258382e-05, | |
| "loss": 1.8704, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 7.546656444541333e-05, | |
| "loss": 1.9435, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 5.2697137691647635e-05, | |
| "loss": 2.0704, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "learning_rate": 3.1687550572992616e-05, | |
| "loss": 2.0372, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 1.4762762169883855e-05, | |
| "loss": 1.9286, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 3.795700315696817e-06, | |
| "loss": 1.7845, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0, | |
| "loss": 1.7819, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.9693368673324585, | |
| "eval_runtime": 1.5591, | |
| "eval_samples_per_second": 42.973, | |
| "eval_steps_per_second": 5.772, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 3.7957003156967485e-06, | |
| "loss": 1.6773, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "learning_rate": 1.4762762169883802e-05, | |
| "loss": 1.9066, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "learning_rate": 3.168755057299255e-05, | |
| "loss": 1.7702, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "learning_rate": 5.269713769164743e-05, | |
| "loss": 1.7718, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "learning_rate": 7.546656444541325e-05, | |
| "loss": 1.6923, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 5.64, | |
| "learning_rate": 9.747612470258363e-05, | |
| "loss": 1.8134, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 5.74, | |
| "learning_rate": 0.0001162902005384805, | |
| "loss": 1.6943, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 5.85, | |
| "learning_rate": 0.00012982679213998787, | |
| "loss": 1.7474, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "learning_rate": 0.00013658791568741792, | |
| "loss": 1.664, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.8830074071884155, | |
| "eval_runtime": 1.6028, | |
| "eval_samples_per_second": 41.801, | |
| "eval_steps_per_second": 5.615, | |
| "step": 282 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 0.00013582537296550986, | |
| "loss": 1.7593, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "learning_rate": 0.00012762354833724559, | |
| "loss": 1.68, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "learning_rate": 0.0001128900706169339, | |
| "loss": 1.5629, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 6.38, | |
| "learning_rate": 9.325537321665346e-05, | |
| "loss": 1.5448, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 6.49, | |
| "learning_rate": 7.089226742272658e-05, | |
| "loss": 1.5643, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 6.6, | |
| "learning_rate": 4.827549515124539e-05, | |
| "loss": 1.5585, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 6.7, | |
| "learning_rate": 2.7907869511566348e-05, | |
| "loss": 1.724, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 6.81, | |
| "learning_rate": 1.2043308960555334e-05, | |
| "loss": 1.3896, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "learning_rate": 2.437414591269752e-06, | |
| "loss": 1.4656, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 1.7848814725875854, | |
| "eval_runtime": 1.6359, | |
| "eval_samples_per_second": 40.956, | |
| "eval_steps_per_second": 5.502, | |
| "step": 329 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 1.5319212645167772e-07, | |
| "loss": 1.4519, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 7.13, | |
| "learning_rate": 5.443417773996994e-06, | |
| "loss": 1.5894, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 7.23, | |
| "learning_rate": 1.772266555255011e-05, | |
| "loss": 1.377, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 7.34, | |
| "learning_rate": 3.563209159419346e-05, | |
| "loss": 1.4357, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 7.45, | |
| "learning_rate": 5.718980627559723e-05, | |
| "loss": 1.5269, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 7.55, | |
| "learning_rate": 8.001019372440265e-05, | |
| "loss": 1.4501, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 7.66, | |
| "learning_rate": 0.00010156790840580641, | |
| "loss": 1.2612, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 7.77, | |
| "learning_rate": 0.0001194773344474498, | |
| "loss": 1.3453, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 7.87, | |
| "learning_rate": 0.00013175658222600294, | |
| "loss": 1.4728, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 7.98, | |
| "learning_rate": 0.00013704680787354832, | |
| "loss": 1.4687, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.76621675491333, | |
| "eval_runtime": 1.6517, | |
| "eval_samples_per_second": 40.565, | |
| "eval_steps_per_second": 5.449, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 8.09, | |
| "learning_rate": 0.00013476258540873022, | |
| "loss": 1.4856, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 8.19, | |
| "learning_rate": 0.00012515669103944473, | |
| "loss": 1.2361, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 8.3, | |
| "learning_rate": 0.00010929213048843395, | |
| "loss": 1.1886, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "learning_rate": 8.892450484875472e-05, | |
| "loss": 1.3658, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 8.51, | |
| "learning_rate": 6.630773257727356e-05, | |
| "loss": 1.3489, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 8.62, | |
| "learning_rate": 4.394462678334666e-05, | |
| "loss": 1.2336, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 8.72, | |
| "learning_rate": 2.4309929383066207e-05, | |
| "loss": 1.2826, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 8.83, | |
| "learning_rate": 9.576451662754362e-06, | |
| "loss": 1.381, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 8.94, | |
| "learning_rate": 1.3746270344902175e-06, | |
| "loss": 1.2151, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 1.6784825325012207, | |
| "eval_runtime": 1.678, | |
| "eval_samples_per_second": 39.927, | |
| "eval_steps_per_second": 5.363, | |
| "step": 423 | |
| } | |
| ], | |
| "max_steps": 470, | |
| "num_train_epochs": 10, | |
| "total_flos": 439754489856000.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |