| { | |
| "best_metric": 0.05191322788596153, | |
| "best_model_checkpoint": "./vit-base-beans/checkpoint-400", | |
| "epoch": 7.142857142857143, | |
| "global_step": 400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00019553571428571428, | |
| "loss": 1.6254, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00019107142857142858, | |
| "loss": 1.0629, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.00018660714285714287, | |
| "loss": 0.7167, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.00018214285714285714, | |
| "loss": 0.4412, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.00017767857142857141, | |
| "loss": 0.3637, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 0.00017321428571428574, | |
| "loss": 0.2357, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.00016875, | |
| "loss": 0.1346, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.00016428571428571428, | |
| "loss": 0.1158, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 0.00015982142857142858, | |
| "loss": 0.0805, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 0.00015535714285714287, | |
| "loss": 0.0835, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "eval_accuracy": 0.967479674796748, | |
| "eval_loss": 0.1520603448152542, | |
| "eval_runtime": 5.265, | |
| "eval_samples_per_second": 46.723, | |
| "eval_steps_per_second": 5.888, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 0.00015089285714285714, | |
| "loss": 0.1141, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 0.00014642857142857141, | |
| "loss": 0.139, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 0.00014196428571428574, | |
| "loss": 0.05, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 0.0001375, | |
| "loss": 0.0443, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 0.00013303571428571428, | |
| "loss": 0.0387, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 0.00012857142857142858, | |
| "loss": 0.0355, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 0.00012410714285714287, | |
| "loss": 0.0609, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 0.00011964285714285714, | |
| "loss": 0.0308, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 0.00011517857142857143, | |
| "loss": 0.0284, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 0.00011071428571428572, | |
| "loss": 0.0265, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "eval_accuracy": 0.991869918699187, | |
| "eval_loss": 0.0650288462638855, | |
| "eval_runtime": 4.3392, | |
| "eval_samples_per_second": 56.692, | |
| "eval_steps_per_second": 7.144, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 0.00010625000000000001, | |
| "loss": 0.0252, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 0.00010178571428571428, | |
| "loss": 0.0234, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 9.732142857142858e-05, | |
| "loss": 0.0221, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "learning_rate": 9.285714285714286e-05, | |
| "loss": 0.0213, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 8.839285714285714e-05, | |
| "loss": 0.0205, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "learning_rate": 8.392857142857144e-05, | |
| "loss": 0.0196, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "learning_rate": 7.946428571428571e-05, | |
| "loss": 0.0188, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0183, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 5.18, | |
| "learning_rate": 7.053571428571429e-05, | |
| "loss": 0.0176, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "learning_rate": 6.607142857142857e-05, | |
| "loss": 0.0171, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "eval_accuracy": 0.991869918699187, | |
| "eval_loss": 0.05430033057928085, | |
| "eval_runtime": 4.4865, | |
| "eval_samples_per_second": 54.831, | |
| "eval_steps_per_second": 6.91, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "learning_rate": 6.160714285714286e-05, | |
| "loss": 0.0165, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 5.71, | |
| "learning_rate": 5.714285714285714e-05, | |
| "loss": 0.016, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 5.89, | |
| "learning_rate": 5.267857142857143e-05, | |
| "loss": 0.0159, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 4.8214285714285716e-05, | |
| "loss": 0.0155, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "learning_rate": 4.375e-05, | |
| "loss": 0.0151, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "learning_rate": 3.928571428571429e-05, | |
| "loss": 0.0149, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 6.61, | |
| "learning_rate": 3.4821428571428574e-05, | |
| "loss": 0.0146, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 6.79, | |
| "learning_rate": 3.0357142857142857e-05, | |
| "loss": 0.0144, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "learning_rate": 2.5892857142857148e-05, | |
| "loss": 0.0142, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "learning_rate": 2.1428571428571428e-05, | |
| "loss": 0.0142, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "eval_accuracy": 0.991869918699187, | |
| "eval_loss": 0.05191322788596153, | |
| "eval_runtime": 4.4319, | |
| "eval_samples_per_second": 55.507, | |
| "eval_steps_per_second": 6.995, | |
| "step": 400 | |
| } | |
| ], | |
| "max_steps": 448, | |
| "num_train_epochs": 8, | |
| "total_flos": 4.894613398587064e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |