|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 5.0, |
|
"global_step": 760, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.07, |
|
"learning_rate": 1.8842105263157898e-05, |
|
"loss": 0.8301, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.13, |
|
"learning_rate": 1.7789473684210527e-05, |
|
"loss": 0.7976, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"learning_rate": 1.673684210526316e-05, |
|
"loss": 0.5757, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.26, |
|
"learning_rate": 1.568421052631579e-05, |
|
"loss": 0.4569, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.33, |
|
"learning_rate": 1.4631578947368424e-05, |
|
"loss": 0.7308, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.39, |
|
"learning_rate": 1.3578947368421055e-05, |
|
"loss": 0.6149, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.46, |
|
"learning_rate": 1.2526315789473684e-05, |
|
"loss": 0.666, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.53, |
|
"learning_rate": 1.1473684210526317e-05, |
|
"loss": 0.6656, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.59, |
|
"learning_rate": 1.0421052631578948e-05, |
|
"loss": 0.5629, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 0.66, |
|
"learning_rate": 9.36842105263158e-06, |
|
"loss": 0.8803, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.72, |
|
"learning_rate": 8.315789473684212e-06, |
|
"loss": 0.4404, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 0.79, |
|
"learning_rate": 7.263157894736843e-06, |
|
"loss": 0.3989, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 0.86, |
|
"learning_rate": 6.2105263157894745e-06, |
|
"loss": 0.5877, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 0.92, |
|
"learning_rate": 5.157894736842106e-06, |
|
"loss": 0.5024, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.99, |
|
"learning_rate": 4.105263157894737e-06, |
|
"loss": 0.7384, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.7962962962962963, |
|
"eval_loss": 0.42684096097946167, |
|
"eval_runtime": 5.5879, |
|
"eval_samples_per_second": 9.664, |
|
"eval_steps_per_second": 1.253, |
|
"step": 152 |
|
}, |
|
{ |
|
"epoch": 1.05, |
|
"learning_rate": 3.052631578947369e-06, |
|
"loss": 0.4284, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 1.12, |
|
"learning_rate": 2.0000000000000003e-06, |
|
"loss": 0.1814, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 1.18, |
|
"learning_rate": 9.473684210526317e-07, |
|
"loss": 0.2693, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 1.25, |
|
"learning_rate": 0.0, |
|
"loss": 0.3471, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 1.32, |
|
"learning_rate": 0.0, |
|
"loss": 0.3491, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 1.38, |
|
"learning_rate": 0.0, |
|
"loss": 0.2447, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 1.45, |
|
"learning_rate": 0.0, |
|
"loss": 0.4893, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 1.51, |
|
"learning_rate": 0.0, |
|
"loss": 0.2365, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 1.58, |
|
"learning_rate": 0.0, |
|
"loss": 0.2919, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 1.64, |
|
"learning_rate": 0.0, |
|
"loss": 0.4932, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 1.71, |
|
"learning_rate": 0.0, |
|
"loss": 0.2127, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 1.78, |
|
"learning_rate": 0.0, |
|
"loss": 0.2355, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 1.84, |
|
"learning_rate": 0.0, |
|
"loss": 0.4058, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 1.91, |
|
"learning_rate": 0.0, |
|
"loss": 0.4578, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 1.97, |
|
"learning_rate": 0.0, |
|
"loss": 0.2888, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.7777777777777778, |
|
"eval_loss": 0.4249849319458008, |
|
"eval_runtime": 5.3688, |
|
"eval_samples_per_second": 10.058, |
|
"eval_steps_per_second": 1.304, |
|
"step": 304 |
|
}, |
|
{ |
|
"epoch": 2.04, |
|
"learning_rate": 0.0, |
|
"loss": 0.5527, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 2.11, |
|
"learning_rate": 0.0, |
|
"loss": 0.4907, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 2.17, |
|
"learning_rate": 0.0, |
|
"loss": 0.6212, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 2.24, |
|
"learning_rate": 0.0, |
|
"loss": 0.3302, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 2.3, |
|
"learning_rate": 0.0, |
|
"loss": 0.4032, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 2.37, |
|
"learning_rate": 0.0, |
|
"loss": 0.2037, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 2.43, |
|
"learning_rate": 0.0, |
|
"loss": 0.2949, |
|
"step": 370 |
|
}, |
|
{ |
|
"epoch": 2.5, |
|
"learning_rate": 0.0, |
|
"loss": 0.3068, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 2.57, |
|
"learning_rate": 0.0, |
|
"loss": 0.2455, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 2.63, |
|
"learning_rate": 0.0, |
|
"loss": 0.2118, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 2.7, |
|
"learning_rate": 0.0, |
|
"loss": 0.5577, |
|
"step": 410 |
|
}, |
|
{ |
|
"epoch": 2.76, |
|
"learning_rate": 0.0, |
|
"loss": 0.3432, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 2.83, |
|
"learning_rate": 0.0, |
|
"loss": 0.2259, |
|
"step": 430 |
|
}, |
|
{ |
|
"epoch": 2.89, |
|
"learning_rate": 0.0, |
|
"loss": 0.2736, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 2.96, |
|
"learning_rate": 0.0, |
|
"loss": 0.2953, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.7777777777777778, |
|
"eval_loss": 0.4249849319458008, |
|
"eval_runtime": 5.4342, |
|
"eval_samples_per_second": 9.937, |
|
"eval_steps_per_second": 1.288, |
|
"step": 456 |
|
}, |
|
{ |
|
"epoch": 3.03, |
|
"learning_rate": 0.0, |
|
"loss": 0.2847, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 3.09, |
|
"learning_rate": 0.0, |
|
"loss": 0.2117, |
|
"step": 470 |
|
}, |
|
{ |
|
"epoch": 3.16, |
|
"learning_rate": 0.0, |
|
"loss": 0.2116, |
|
"step": 480 |
|
}, |
|
{ |
|
"epoch": 3.22, |
|
"learning_rate": 0.0, |
|
"loss": 0.3803, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 3.29, |
|
"learning_rate": 0.0, |
|
"loss": 0.3206, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 3.36, |
|
"learning_rate": 0.0, |
|
"loss": 0.4362, |
|
"step": 510 |
|
}, |
|
{ |
|
"epoch": 3.42, |
|
"learning_rate": 0.0, |
|
"loss": 0.4031, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 3.49, |
|
"learning_rate": 0.0, |
|
"loss": 0.2682, |
|
"step": 530 |
|
}, |
|
{ |
|
"epoch": 3.55, |
|
"learning_rate": 0.0, |
|
"loss": 0.292, |
|
"step": 540 |
|
}, |
|
{ |
|
"epoch": 3.62, |
|
"learning_rate": 0.0, |
|
"loss": 0.3007, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 3.68, |
|
"learning_rate": 0.0, |
|
"loss": 0.5192, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 3.75, |
|
"learning_rate": 0.0, |
|
"loss": 0.3809, |
|
"step": 570 |
|
}, |
|
{ |
|
"epoch": 3.82, |
|
"learning_rate": 0.0, |
|
"loss": 0.3756, |
|
"step": 580 |
|
}, |
|
{ |
|
"epoch": 3.88, |
|
"learning_rate": 0.0, |
|
"loss": 0.3174, |
|
"step": 590 |
|
}, |
|
{ |
|
"epoch": 3.95, |
|
"learning_rate": 0.0, |
|
"loss": 0.4914, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.7777777777777778, |
|
"eval_loss": 0.4249849319458008, |
|
"eval_runtime": 5.3734, |
|
"eval_samples_per_second": 10.05, |
|
"eval_steps_per_second": 1.303, |
|
"step": 608 |
|
}, |
|
{ |
|
"epoch": 4.01, |
|
"learning_rate": 0.0, |
|
"loss": 0.2904, |
|
"step": 610 |
|
}, |
|
{ |
|
"epoch": 4.08, |
|
"learning_rate": 0.0, |
|
"loss": 0.433, |
|
"step": 620 |
|
}, |
|
{ |
|
"epoch": 4.14, |
|
"learning_rate": 0.0, |
|
"loss": 0.1684, |
|
"step": 630 |
|
}, |
|
{ |
|
"epoch": 4.21, |
|
"learning_rate": 0.0, |
|
"loss": 0.2903, |
|
"step": 640 |
|
}, |
|
{ |
|
"epoch": 4.28, |
|
"learning_rate": 0.0, |
|
"loss": 0.3659, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 4.34, |
|
"learning_rate": 0.0, |
|
"loss": 0.2125, |
|
"step": 660 |
|
}, |
|
{ |
|
"epoch": 4.41, |
|
"learning_rate": 0.0, |
|
"loss": 0.3639, |
|
"step": 670 |
|
}, |
|
{ |
|
"epoch": 4.47, |
|
"learning_rate": 0.0, |
|
"loss": 0.4092, |
|
"step": 680 |
|
}, |
|
{ |
|
"epoch": 4.54, |
|
"learning_rate": 0.0, |
|
"loss": 0.3467, |
|
"step": 690 |
|
}, |
|
{ |
|
"epoch": 4.61, |
|
"learning_rate": 0.0, |
|
"loss": 0.3013, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 4.67, |
|
"learning_rate": 0.0, |
|
"loss": 0.2546, |
|
"step": 710 |
|
}, |
|
{ |
|
"epoch": 4.74, |
|
"learning_rate": 0.0, |
|
"loss": 0.3741, |
|
"step": 720 |
|
}, |
|
{ |
|
"epoch": 4.8, |
|
"learning_rate": 0.0, |
|
"loss": 0.394, |
|
"step": 730 |
|
}, |
|
{ |
|
"epoch": 4.87, |
|
"learning_rate": 0.0, |
|
"loss": 0.2283, |
|
"step": 740 |
|
}, |
|
{ |
|
"epoch": 4.93, |
|
"learning_rate": 0.0, |
|
"loss": 0.4307, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"learning_rate": 0.0, |
|
"loss": 0.4099, |
|
"step": 760 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_accuracy": 0.7777777777777778, |
|
"eval_loss": 0.4249849319458008, |
|
"eval_runtime": 5.8483, |
|
"eval_samples_per_second": 9.233, |
|
"eval_steps_per_second": 1.197, |
|
"step": 760 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"step": 760, |
|
"total_flos": 7.945332390564987e+17, |
|
"train_loss": 0.39741649894337905, |
|
"train_runtime": 401.1562, |
|
"train_samples_per_second": 3.789, |
|
"train_steps_per_second": 1.895 |
|
} |
|
], |
|
"max_steps": 760, |
|
"num_train_epochs": 5, |
|
"total_flos": 7.945332390564987e+17, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|