| { |
| "best_metric": 0.13027848303318024, |
| "best_model_checkpoint": "vit_no_aug_1e5/checkpoint-4500", |
| "epoch": 3.0, |
| "eval_steps": 500, |
| "global_step": 4500, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.3333333333333333, |
| "grad_norm": 9.59843921661377, |
| "learning_rate": 8.988764044943822e-06, |
| "loss": 0.2951, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.6666666666666666, |
| "grad_norm": 1.7021756172180176, |
| "learning_rate": 7.865168539325843e-06, |
| "loss": 0.1504, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 2.5123584270477295, |
| "learning_rate": 6.741573033707865e-06, |
| "loss": 0.1296, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.929625, |
| "eval_loss": 0.19084413349628448, |
| "eval_model_preparation_time": 0.0048, |
| "eval_runtime": 193.8847, |
| "eval_samples_per_second": 123.785, |
| "eval_steps_per_second": 3.868, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.3333333333333333, |
| "grad_norm": 1.9735125303268433, |
| "learning_rate": 5.617977528089889e-06, |
| "loss": 0.1107, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.6666666666666665, |
| "grad_norm": 2.696586847305298, |
| "learning_rate": 4.494382022471911e-06, |
| "loss": 0.0999, |
| "step": 2500 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 1.1763291358947754, |
| "learning_rate": 3.3707865168539327e-06, |
| "loss": 0.095, |
| "step": 3000 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.942375, |
| "eval_loss": 0.15168513357639313, |
| "eval_model_preparation_time": 0.0048, |
| "eval_runtime": 199.5805, |
| "eval_samples_per_second": 120.252, |
| "eval_steps_per_second": 3.758, |
| "step": 3000 |
| }, |
| { |
| "epoch": 2.3333333333333335, |
| "grad_norm": 1.1426674127578735, |
| "learning_rate": 2.2471910112359554e-06, |
| "loss": 0.0845, |
| "step": 3500 |
| }, |
| { |
| "epoch": 2.6666666666666665, |
| "grad_norm": 4.490411281585693, |
| "learning_rate": 1.1235955056179777e-06, |
| "loss": 0.0854, |
| "step": 4000 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 1.7655569314956665, |
| "learning_rate": 0.0, |
| "loss": 0.0758, |
| "step": 4500 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.95125, |
| "eval_loss": 0.13027848303318024, |
| "eval_model_preparation_time": 0.0048, |
| "eval_runtime": 194.2224, |
| "eval_samples_per_second": 123.57, |
| "eval_steps_per_second": 3.862, |
| "step": 4500 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 4500, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.2317693009002496e+19, |
| "train_batch_size": 64, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
