| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "global_step": 5268, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 1.810174639331815e-05, | |
| "loss": 0.2574, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 1.6203492786636296e-05, | |
| "loss": 0.1076, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 1.4305239179954442e-05, | |
| "loss": 0.0794, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9808382881026667, | |
| "eval_f1": 0.9228973910144204, | |
| "eval_loss": 0.07026076316833496, | |
| "eval_precision": 0.9142857142857143, | |
| "eval_recall": 0.9316728374284753, | |
| "eval_runtime": 490.5826, | |
| "eval_samples_per_second": 6.625, | |
| "eval_steps_per_second": 0.83, | |
| "step": 1756 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 1.240698557327259e-05, | |
| "loss": 0.0588, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 1.0508731966590738e-05, | |
| "loss": 0.0419, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 8.610478359908885e-06, | |
| "loss": 0.0437, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 6.712224753227031e-06, | |
| "loss": 0.0346, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9861217401542356, | |
| "eval_f1": 0.9410095953274927, | |
| "eval_loss": 0.057285286486148834, | |
| "eval_precision": 0.9331457885156379, | |
| "eval_recall": 0.9490070683271625, | |
| "eval_runtime": 491.5796, | |
| "eval_samples_per_second": 6.611, | |
| "eval_steps_per_second": 0.828, | |
| "step": 3512 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 4.8139711465451785e-06, | |
| "loss": 0.0258, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 2.9157175398633257e-06, | |
| "loss": 0.0191, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 1.0174639331814731e-06, | |
| "loss": 0.0191, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.986504385706717, | |
| "eval_f1": 0.9437489567684861, | |
| "eval_loss": 0.05966141074895859, | |
| "eval_precision": 0.936092715231788, | |
| "eval_recall": 0.9515314708852238, | |
| "eval_runtime": 487.1838, | |
| "eval_samples_per_second": 6.671, | |
| "eval_steps_per_second": 0.835, | |
| "step": 5268 | |
| } | |
| ], | |
| "max_steps": 5268, | |
| "num_train_epochs": 3, | |
| "total_flos": 923635312577460.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |