| { | |
| "best_metric": 0.12367749214172363, | |
| "best_model_checkpoint": "./vit-base-beans/checkpoint-300", | |
| "epoch": 1.8072289156626506, | |
| "global_step": 300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00019951807228915663, | |
| "loss": 1.6515, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00019903614457831325, | |
| "loss": 1.0551, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0001985542168674699, | |
| "loss": 0.6568, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.00019807228915662652, | |
| "loss": 0.4811, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00019759036144578314, | |
| "loss": 0.3576, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00019710843373493977, | |
| "loss": 0.2787, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00019662650602409642, | |
| "loss": 0.233, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.000196144578313253, | |
| "loss": 0.1287, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.00019566265060240966, | |
| "loss": 0.1755, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00019518072289156628, | |
| "loss": 0.109, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_accuracy": 0.959349593495935, | |
| "eval_loss": 0.16569873690605164, | |
| "eval_runtime": 4.541, | |
| "eval_samples_per_second": 54.173, | |
| "eval_steps_per_second": 6.827, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0001946987951807229, | |
| "loss": 0.1129, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00019421686746987952, | |
| "loss": 0.1769, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.00019373493975903617, | |
| "loss": 0.1511, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.00019325301204819277, | |
| "loss": 0.2121, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.00019277108433734942, | |
| "loss": 0.135, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.00019228915662650604, | |
| "loss": 0.1046, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.00019180722891566266, | |
| "loss": 0.0742, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.00019132530120481928, | |
| "loss": 0.0655, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 0.00019084337349397593, | |
| "loss": 0.0735, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.00019036144578313252, | |
| "loss": 0.0599, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_accuracy": 0.9471544715447154, | |
| "eval_loss": 0.16865180432796478, | |
| "eval_runtime": 4.123, | |
| "eval_samples_per_second": 59.665, | |
| "eval_steps_per_second": 7.519, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 0.00018987951807228917, | |
| "loss": 0.1547, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.0001893975903614458, | |
| "loss": 0.0723, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 0.00018891566265060242, | |
| "loss": 0.045, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.00018843373493975904, | |
| "loss": 0.0339, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 0.00018795180722891569, | |
| "loss": 0.0424, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 0.00018746987951807228, | |
| "loss": 0.1061, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 0.00018698795180722893, | |
| "loss": 0.0557, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.00018650602409638555, | |
| "loss": 0.0491, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.00018602409638554217, | |
| "loss": 0.0489, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 0.0001855421686746988, | |
| "loss": 0.0387, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "eval_accuracy": 0.9715447154471545, | |
| "eval_loss": 0.12367749214172363, | |
| "eval_runtime": 4.2743, | |
| "eval_samples_per_second": 57.553, | |
| "eval_steps_per_second": 7.253, | |
| "step": 300 | |
| } | |
| ], | |
| "max_steps": 4150, | |
| "num_train_epochs": 25, | |
| "total_flos": 3.716682371694674e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |