| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 1250, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0005, | |
| "loss": 2.3712, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0005, | |
| "loss": 1.8732, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.593873417721519, | |
| "eval_loss": 2.011091947555542, | |
| "eval_runtime": 4.861, | |
| "eval_samples_per_second": 102.859, | |
| "eval_steps_per_second": 12.96, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_exact_match": 0.0, | |
| "eval_f1": 0.32835692091264185, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.0005, | |
| "loss": 1.7329, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0005, | |
| "loss": 1.6233, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0005, | |
| "loss": 1.6142, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6051139240506329, | |
| "eval_loss": 1.844284176826477, | |
| "eval_runtime": 4.8205, | |
| "eval_samples_per_second": 103.724, | |
| "eval_steps_per_second": 13.069, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_exact_match": 0.0, | |
| "eval_f1": 0.7215873015873017, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0005, | |
| "loss": 1.1625, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 0.0005, | |
| "loss": 1.206, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6007341772151898, | |
| "eval_loss": 1.981817603111267, | |
| "eval_runtime": 4.671, | |
| "eval_samples_per_second": 107.044, | |
| "eval_steps_per_second": 13.487, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_exact_match": 0.0, | |
| "eval_f1": 0.4163636363636364, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 0.0005, | |
| "loss": 1.0064, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 0.0005, | |
| "loss": 0.8141, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0005, | |
| "loss": 0.8693, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.594126582278481, | |
| "eval_loss": 2.2100281715393066, | |
| "eval_runtime": 5.2488, | |
| "eval_samples_per_second": 95.26, | |
| "eval_steps_per_second": 12.003, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_exact_match": 0.2, | |
| "eval_f1": 1.559097642686839, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "learning_rate": 0.0005, | |
| "loss": 0.5477, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 0.0005, | |
| "loss": 0.6023, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5910126582278481, | |
| "eval_loss": 2.375615119934082, | |
| "eval_runtime": 4.4156, | |
| "eval_samples_per_second": 113.235, | |
| "eval_steps_per_second": 14.268, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_exact_match": 0.4, | |
| "eval_f1": 2.41971862073413, | |
| "step": 1250 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 12500, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 7.98552666144768e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |