| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.0, | |
| "global_step": 96, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0914, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0328, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_acc": 0.746268630027771, | |
| "eval_loss": 0.9441611170768738, | |
| "eval_runtime": 0.857, | |
| "eval_samples_per_second": 78.18, | |
| "eval_steps_per_second": 5.834, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9558, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8742, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_acc": 0.9402984976768494, | |
| "eval_loss": 0.7099391222000122, | |
| "eval_runtime": 0.8562, | |
| "eval_samples_per_second": 78.255, | |
| "eval_steps_per_second": 5.84, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7474, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6858, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6451, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_acc": 0.9402984976768494, | |
| "eval_loss": 0.5049594044685364, | |
| "eval_runtime": 0.8709, | |
| "eval_samples_per_second": 76.935, | |
| "eval_steps_per_second": 5.741, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5587, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 2e-05, | |
| "loss": 0.508, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_acc": 0.9402984976768494, | |
| "eval_loss": 0.37610408663749695, | |
| "eval_runtime": 0.8617, | |
| "eval_samples_per_second": 77.756, | |
| "eval_steps_per_second": 5.803, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 96, | |
| "total_flos": 0.0, | |
| "train_loss": 0.7672463009754816, | |
| "train_runtime": 42.5944, | |
| "train_samples_per_second": 35.779, | |
| "train_steps_per_second": 2.254 | |
| } | |
| ], | |
| "max_steps": 96, | |
| "num_train_epochs": 4, | |
| "total_flos": 0.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |