| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.62962962962963, | |
| "global_step": 5500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 4.57912457912458e-05, | |
| "loss": 0.2927, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 4.158249158249159e-05, | |
| "loss": 0.1893, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.962, | |
| "eval_f1": 0.9604989604989606, | |
| "eval_loss": 0.14235617220401764, | |
| "eval_runtime": 1.4452, | |
| "eval_samples_per_second": 345.973, | |
| "eval_steps_per_second": 43.593, | |
| "step": 1188 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 3.7373737373737376e-05, | |
| "loss": 0.1151, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 3.3164983164983165e-05, | |
| "loss": 0.0722, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.956, | |
| "eval_f1": 0.9539748953974895, | |
| "eval_loss": 0.2519678771495819, | |
| "eval_runtime": 1.3777, | |
| "eval_samples_per_second": 362.913, | |
| "eval_steps_per_second": 45.727, | |
| "step": 2376 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 2.8956228956228958e-05, | |
| "loss": 0.0563, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 2.474747474747475e-05, | |
| "loss": 0.0135, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 2.0538720538720542e-05, | |
| "loss": 0.0204, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.962, | |
| "eval_f1": 0.9608247422680413, | |
| "eval_loss": 0.2846008539199829, | |
| "eval_runtime": 1.2758, | |
| "eval_samples_per_second": 391.904, | |
| "eval_steps_per_second": 49.38, | |
| "step": 3564 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 1.632996632996633e-05, | |
| "loss": 0.005, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 1.2121212121212122e-05, | |
| "loss": 0.0001, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.956, | |
| "eval_f1": 0.9541666666666667, | |
| "eval_loss": 0.3692273795604706, | |
| "eval_runtime": 1.4564, | |
| "eval_samples_per_second": 343.322, | |
| "eval_steps_per_second": 43.259, | |
| "step": 4752 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 7.912457912457913e-06, | |
| "loss": 0.0024, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.0, | |
| "step": 5500 | |
| } | |
| ], | |
| "max_steps": 5940, | |
| "num_train_epochs": 5, | |
| "total_flos": 555135681114528.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |