| { | |
| "best_metric": 0.006926598027348518, | |
| "best_model_checkpoint": "./outputs/checkpoint-605", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 605, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.9669421487603307e-05, | |
| "loss": 1.3579, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 1.9338842975206613e-05, | |
| "loss": 1.2345, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 1.900826446280992e-05, | |
| "loss": 1.1497, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 1.8677685950413225e-05, | |
| "loss": 1.0313, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 1.834710743801653e-05, | |
| "loss": 0.9598, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 1.8016528925619837e-05, | |
| "loss": 0.802, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 1.7685950413223143e-05, | |
| "loss": 0.7353, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 1.735537190082645e-05, | |
| "loss": 0.6321, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 1.7024793388429754e-05, | |
| "loss": 0.5063, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 1.669421487603306e-05, | |
| "loss": 0.4158, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 1.6363636363636366e-05, | |
| "loss": 0.3683, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 1.6033057851239672e-05, | |
| "loss": 0.2849, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.15511935949325562, | |
| "eval_runtime": 8.1016, | |
| "eval_samples_per_second": 118.866, | |
| "eval_steps_per_second": 7.529, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 1.5702479338842978e-05, | |
| "loss": 0.2395, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 1.5371900826446283e-05, | |
| "loss": 0.1841, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 1.504132231404959e-05, | |
| "loss": 0.1696, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 1.4710743801652893e-05, | |
| "loss": 0.1363, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 1.4380165289256201e-05, | |
| "loss": 0.103, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 1.4049586776859505e-05, | |
| "loss": 0.104, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.3719008264462813e-05, | |
| "loss": 0.1042, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 1.3388429752066117e-05, | |
| "loss": 0.0718, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 1.3057851239669424e-05, | |
| "loss": 0.0798, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 1.2727272727272728e-05, | |
| "loss": 0.0758, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 1.2396694214876034e-05, | |
| "loss": 0.0589, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 1.206611570247934e-05, | |
| "loss": 0.0734, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.01950172521173954, | |
| "eval_runtime": 8.0335, | |
| "eval_samples_per_second": 119.873, | |
| "eval_steps_per_second": 7.593, | |
| "step": 242 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 1.1735537190082646e-05, | |
| "loss": 0.0698, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 1.1404958677685952e-05, | |
| "loss": 0.0705, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 1.1074380165289258e-05, | |
| "loss": 0.0522, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 1.0743801652892562e-05, | |
| "loss": 0.0645, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 1.041322314049587e-05, | |
| "loss": 0.0433, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 1.0082644628099174e-05, | |
| "loss": 0.0382, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 9.75206611570248e-06, | |
| "loss": 0.0583, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 9.421487603305785e-06, | |
| "loss": 0.0435, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 0.0377, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 8.760330578512397e-06, | |
| "loss": 0.0477, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 8.429752066115703e-06, | |
| "loss": 0.0443, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 8.099173553719009e-06, | |
| "loss": 0.0412, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.009915151633322239, | |
| "eval_runtime": 8.1318, | |
| "eval_samples_per_second": 118.423, | |
| "eval_steps_per_second": 7.501, | |
| "step": 363 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 7.768595041322314e-06, | |
| "loss": 0.0295, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 7.43801652892562e-06, | |
| "loss": 0.025, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 7.107438016528926e-06, | |
| "loss": 0.047, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 6.776859504132232e-06, | |
| "loss": 0.0375, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 6.446280991735537e-06, | |
| "loss": 0.035, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 6.115702479338843e-06, | |
| "loss": 0.0241, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 5.785123966942149e-06, | |
| "loss": 0.0253, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 5.4545454545454545e-06, | |
| "loss": 0.0458, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 5.12396694214876e-06, | |
| "loss": 0.0241, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 4.793388429752067e-06, | |
| "loss": 0.0267, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 4.462809917355372e-06, | |
| "loss": 0.0324, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "learning_rate": 4.132231404958678e-06, | |
| "loss": 0.0457, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.007458301726728678, | |
| "eval_runtime": 8.0474, | |
| "eval_samples_per_second": 119.667, | |
| "eval_steps_per_second": 7.58, | |
| "step": 484 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 3.801652892561984e-06, | |
| "loss": 0.031, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 3.4710743801652895e-06, | |
| "loss": 0.0622, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 3.1404958677685953e-06, | |
| "loss": 0.0399, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "learning_rate": 2.809917355371901e-06, | |
| "loss": 0.0287, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "learning_rate": 2.479338842975207e-06, | |
| "loss": 0.0363, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 2.1487603305785124e-06, | |
| "loss": 0.0484, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "learning_rate": 1.8181818181818183e-06, | |
| "loss": 0.0532, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "learning_rate": 1.4876033057851241e-06, | |
| "loss": 0.046, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "learning_rate": 1.15702479338843e-06, | |
| "loss": 0.016, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 8.264462809917356e-07, | |
| "loss": 0.0331, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "learning_rate": 4.958677685950413e-07, | |
| "loss": 0.0554, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "learning_rate": 1.6528925619834713e-07, | |
| "loss": 0.0342, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.006926598027348518, | |
| "eval_runtime": 8.1069, | |
| "eval_samples_per_second": 118.788, | |
| "eval_steps_per_second": 7.524, | |
| "step": 605 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 605, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 8.594252068818125e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |