| { | |
| "best_metric": 0.006127053406089544, | |
| "best_model_checkpoint": "./vit-base-beans/checkpoint-200", | |
| "epoch": 4.0, | |
| "global_step": 260, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0001930769230769231, | |
| "loss": 2.9271, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0001853846153846154, | |
| "loss": 0.4676, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0001776923076923077, | |
| "loss": 0.3661, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00017, | |
| "loss": 0.2549, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.0001623076923076923, | |
| "loss": 0.229, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.00015461538461538464, | |
| "loss": 0.2858, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.00014692307692307693, | |
| "loss": 0.1084, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.00013923076923076923, | |
| "loss": 0.1062, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.00013153846153846156, | |
| "loss": 0.1174, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 0.00012384615384615385, | |
| "loss": 0.0994, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "eval_accuracy": 0.9849624060150376, | |
| "eval_loss": 0.05971372127532959, | |
| "eval_runtime": 1.0813, | |
| "eval_samples_per_second": 122.997, | |
| "eval_steps_per_second": 15.721, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.00011615384615384617, | |
| "loss": 0.1219, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 0.00010846153846153846, | |
| "loss": 0.0793, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.00010076923076923077, | |
| "loss": 0.0741, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 9.307692307692309e-05, | |
| "loss": 0.0281, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 8.538461538461538e-05, | |
| "loss": 0.0309, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 7.76923076923077e-05, | |
| "loss": 0.1636, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 7e-05, | |
| "loss": 0.0244, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 6.23076923076923e-05, | |
| "loss": 0.0115, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 5.461538461538461e-05, | |
| "loss": 0.0046, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 4.692307692307693e-05, | |
| "loss": 0.001, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.006127053406089544, | |
| "eval_runtime": 1.5958, | |
| "eval_samples_per_second": 83.342, | |
| "eval_steps_per_second": 10.653, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 3.923076923076923e-05, | |
| "loss": 0.0081, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 3.153846153846154e-05, | |
| "loss": 0.0004, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 2.384615384615385e-05, | |
| "loss": 0.0009, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 1.6153846153846154e-05, | |
| "loss": 0.0004, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 8.461538461538462e-06, | |
| "loss": 0.0019, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 7.692307692307694e-07, | |
| "loss": 0.0026, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 260, | |
| "total_flos": 3.233737255326843e+17, | |
| "train_loss": 0.21213595934140567, | |
| "train_runtime": 75.7831, | |
| "train_samples_per_second": 54.577, | |
| "train_steps_per_second": 3.431 | |
| } | |
| ], | |
| "max_steps": 260, | |
| "num_train_epochs": 4, | |
| "total_flos": 3.233737255326843e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |