| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "global_step": 3205, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 4.219968798751951e-05, | |
| "loss": 0.2448, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 3.4399375975039005e-05, | |
| "loss": 0.117, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_accuracy": 0.9510651573223995, | |
| "eval_f1": 0.9413793103448275, | |
| "eval_loss": 0.17917607724666595, | |
| "eval_precision": 0.942101226993865, | |
| "eval_recall": 0.9406584992343032, | |
| "eval_runtime": 1.3355, | |
| "eval_samples_per_second": 548.105, | |
| "eval_steps_per_second": 68.887, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 2.65990639625585e-05, | |
| "loss": 0.0721, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 1.8798751950078e-05, | |
| "loss": 0.045, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "eval_accuracy": 0.9569480345841875, | |
| "eval_f1": 0.9488834696122029, | |
| "eval_loss": 0.20152758061885834, | |
| "eval_precision": 0.948112756808409, | |
| "eval_recall": 0.9496554364471669, | |
| "eval_runtime": 1.2725, | |
| "eval_samples_per_second": 575.238, | |
| "eval_steps_per_second": 72.298, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 1.0998439937597505e-05, | |
| "loss": 0.0252, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "learning_rate": 3.198127925117005e-06, | |
| "loss": 0.0142, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "eval_accuracy": 0.957215438096087, | |
| "eval_f1": 0.94943793350873, | |
| "eval_loss": 0.22939395904541016, | |
| "eval_precision": 0.9490293583245673, | |
| "eval_recall": 0.9498468606431854, | |
| "eval_runtime": 1.2703, | |
| "eval_samples_per_second": 576.223, | |
| "eval_steps_per_second": 72.421, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 3205, | |
| "total_flos": 824467030346496.0, | |
| "train_loss": 0.0817642333912961, | |
| "train_runtime": 184.8416, | |
| "train_samples_per_second": 138.605, | |
| "train_steps_per_second": 17.339 | |
| } | |
| ], | |
| "max_steps": 3205, | |
| "num_train_epochs": 5, | |
| "total_flos": 824467030346496.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |