| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "global_step": 16110, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 5.17063081695967e-06, | |
| "loss": 0.9438, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 9.97820775275705e-06, | |
| "loss": 0.6981, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 9.648022188469921e-06, | |
| "loss": 0.6641, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 9.317836624182791e-06, | |
| "loss": 0.6351, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 8.987651059895663e-06, | |
| "loss": 0.6069, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 8.657465495608533e-06, | |
| "loss": 0.6, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7530310750007629, | |
| "eval_loss": 0.5826240181922913, | |
| "eval_runtime": 74.1302, | |
| "eval_samples_per_second": 132.402, | |
| "eval_steps_per_second": 4.141, | |
| "step": 3222 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 8.327279931321404e-06, | |
| "loss": 0.5475, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 7.997094367034274e-06, | |
| "loss": 0.5118, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 7.666908802747146e-06, | |
| "loss": 0.5113, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 7.336723238460015e-06, | |
| "loss": 0.5134, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 7.006537674172885e-06, | |
| "loss": 0.5022, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 6.676352109885757e-06, | |
| "loss": 0.5082, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7587366104125977, | |
| "eval_loss": 0.5997494459152222, | |
| "eval_runtime": 74.1755, | |
| "eval_samples_per_second": 132.321, | |
| "eval_steps_per_second": 4.139, | |
| "step": 6444 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 6.3461665455986265e-06, | |
| "loss": 0.4847, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 6.015980981311498e-06, | |
| "loss": 0.3947, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 5.685795417024368e-06, | |
| "loss": 0.3968, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 5.3556098527372394e-06, | |
| "loss": 0.3938, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 5.025424288450109e-06, | |
| "loss": 0.3877, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 4.69523872416298e-06, | |
| "loss": 0.3839, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 4.365053159875851e-06, | |
| "loss": 0.3949, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8090677261352539, | |
| "eval_loss": 0.5072706937789917, | |
| "eval_runtime": 74.3802, | |
| "eval_samples_per_second": 131.957, | |
| "eval_steps_per_second": 4.127, | |
| "step": 9666 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 4.034867595588721e-06, | |
| "loss": 0.321, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 3.704682031301592e-06, | |
| "loss": 0.2894, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 3.3744964670144627e-06, | |
| "loss": 0.2923, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 3.0443109027273334e-06, | |
| "loss": 0.2906, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 2.714125338440204e-06, | |
| "loss": 0.2913, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 2.3839397741530743e-06, | |
| "loss": 0.2969, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7917473316192627, | |
| "eval_loss": 0.6656549572944641, | |
| "eval_runtime": 74.3184, | |
| "eval_samples_per_second": 132.067, | |
| "eval_steps_per_second": 4.131, | |
| "step": 12888 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 2.053754209865945e-06, | |
| "loss": 0.2657, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 1.7235686455788153e-06, | |
| "loss": 0.2215, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 1.393383081291686e-06, | |
| "loss": 0.2228, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 1.0631975170045567e-06, | |
| "loss": 0.215, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 4.66, | |
| "learning_rate": 7.330119527174273e-07, | |
| "loss": 0.2306, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "learning_rate": 4.0282638843029786e-07, | |
| "loss": 0.2168, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 7.264082414316847e-08, | |
| "loss": 0.2267, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.801732063293457, | |
| "eval_loss": 0.7047895789146423, | |
| "eval_runtime": 74.3187, | |
| "eval_samples_per_second": 132.066, | |
| "eval_steps_per_second": 4.131, | |
| "step": 16110 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 16110, | |
| "total_flos": 1.2007859696955264e+17, | |
| "train_loss": 0.41926650207727906, | |
| "train_runtime": 12549.4046, | |
| "train_samples_per_second": 41.069, | |
| "train_steps_per_second": 1.284 | |
| } | |
| ], | |
| "max_steps": 16110, | |
| "num_train_epochs": 5, | |
| "total_flos": 1.2007859696955264e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |