| { | |
| "best_metric": 0.9589285714285715, | |
| "best_model_checkpoint": "vit-large-patch16-224-in21k-dungeon-geo-morphs-0-4-30Nov24-003/checkpoint-50", | |
| "epoch": 31.90909090909091, | |
| "eval_steps": 10, | |
| "global_step": 80, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.909090909090909, | |
| "grad_norm": 16.135948181152344, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 1.3682, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 3.909090909090909, | |
| "eval_accuracy": 0.825, | |
| "eval_loss": 1.0376439094543457, | |
| "eval_runtime": 7.5608, | |
| "eval_samples_per_second": 74.066, | |
| "eval_steps_per_second": 9.258, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 7.909090909090909, | |
| "grad_norm": 7.783586025238037, | |
| "learning_rate": 1.6944444444444446e-05, | |
| "loss": 0.5281, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 7.909090909090909, | |
| "eval_accuracy": 0.9196428571428571, | |
| "eval_loss": 0.530949592590332, | |
| "eval_runtime": 7.6225, | |
| "eval_samples_per_second": 73.467, | |
| "eval_steps_per_second": 9.183, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 11.909090909090908, | |
| "grad_norm": 4.47883939743042, | |
| "learning_rate": 1.416666666666667e-05, | |
| "loss": 0.1424, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 11.909090909090908, | |
| "eval_accuracy": 0.9375, | |
| "eval_loss": 0.2757301926612854, | |
| "eval_runtime": 8.417, | |
| "eval_samples_per_second": 66.532, | |
| "eval_steps_per_second": 8.316, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 15.909090909090908, | |
| "grad_norm": 0.6547029614448547, | |
| "learning_rate": 1.138888888888889e-05, | |
| "loss": 0.033, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 15.909090909090908, | |
| "eval_accuracy": 0.9482142857142857, | |
| "eval_loss": 0.16814185678958893, | |
| "eval_runtime": 7.3224, | |
| "eval_samples_per_second": 76.478, | |
| "eval_steps_per_second": 9.56, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 19.90909090909091, | |
| "grad_norm": 0.3083936274051666, | |
| "learning_rate": 8.611111111111112e-06, | |
| "loss": 0.0093, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 19.90909090909091, | |
| "eval_accuracy": 0.9589285714285715, | |
| "eval_loss": 0.13735760748386383, | |
| "eval_runtime": 7.5693, | |
| "eval_samples_per_second": 73.983, | |
| "eval_steps_per_second": 9.248, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 23.90909090909091, | |
| "grad_norm": 0.14518193900585175, | |
| "learning_rate": 5.833333333333334e-06, | |
| "loss": 0.0046, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 23.90909090909091, | |
| "eval_accuracy": 0.9589285714285715, | |
| "eval_loss": 0.12884944677352905, | |
| "eval_runtime": 7.9661, | |
| "eval_samples_per_second": 70.298, | |
| "eval_steps_per_second": 8.787, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 27.90909090909091, | |
| "grad_norm": 0.10925909131765366, | |
| "learning_rate": 3.055555555555556e-06, | |
| "loss": 0.0034, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 27.90909090909091, | |
| "eval_accuracy": 0.9571428571428572, | |
| "eval_loss": 0.12211893498897552, | |
| "eval_runtime": 7.4941, | |
| "eval_samples_per_second": 74.725, | |
| "eval_steps_per_second": 9.341, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 31.90909090909091, | |
| "grad_norm": 0.09304070472717285, | |
| "learning_rate": 2.7777777777777776e-07, | |
| "loss": 0.003, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 31.90909090909091, | |
| "eval_accuracy": 0.9571428571428572, | |
| "eval_loss": 0.12084805965423584, | |
| "eval_runtime": 7.2756, | |
| "eval_samples_per_second": 76.97, | |
| "eval_steps_per_second": 9.621, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 31.90909090909091, | |
| "step": 80, | |
| "total_flos": 7.012786101918106e+17, | |
| "train_loss": 0.2615024119615555, | |
| "train_runtime": 561.0998, | |
| "train_samples_per_second": 5.846, | |
| "train_steps_per_second": 0.143 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 80, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 40, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.012786101918106e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |