| { | |
| "best_metric": 0.27383747696876526, | |
| "best_model_checkpoint": "models/roberta-teacher/checkpoint-8346", | |
| "epoch": 4.0, | |
| "global_step": 16692, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 1.952072849269111e-05, | |
| "loss": 0.4076, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1.904145698538222e-05, | |
| "loss": 0.3484, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 1.856218547807333e-05, | |
| "loss": 0.3104, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 1.808291397076444e-05, | |
| "loss": 0.3241, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 1.7603642463455548e-05, | |
| "loss": 0.31, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1.712437095614666e-05, | |
| "loss": 0.3088, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 1.6645099448837766e-05, | |
| "loss": 0.303, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 1.616582794152888e-05, | |
| "loss": 0.3046, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8758252121974222, | |
| "eval_f1": 0.8723680200515267, | |
| "eval_kappa": 0.5282754100007345, | |
| "eval_loss": 0.28437086939811707, | |
| "eval_precision": 0.8700935888757716, | |
| "eval_recall": 0.8758252121974222, | |
| "eval_runtime": 101.4778, | |
| "eval_samples_per_second": 94.04, | |
| "step": 4173 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 1.5686556434219985e-05, | |
| "loss": 0.2827, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 1.5207284926911097e-05, | |
| "loss": 0.2861, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 1.4728013419602205e-05, | |
| "loss": 0.3014, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 1.4248741912293316e-05, | |
| "loss": 0.2807, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 1.3769470404984425e-05, | |
| "loss": 0.2924, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 1.3290198897675534e-05, | |
| "loss": 0.3189, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 1.2810927390366643e-05, | |
| "loss": 0.2973, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 1.2331655883057754e-05, | |
| "loss": 0.266, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8917531174683013, | |
| "eval_f1": 0.8844462847060539, | |
| "eval_kappa": 0.560786854721061, | |
| "eval_loss": 0.27383747696876526, | |
| "eval_precision": 0.8838724313266608, | |
| "eval_recall": 0.8917531174683013, | |
| "eval_runtime": 100.1991, | |
| "eval_samples_per_second": 95.24, | |
| "step": 8346 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 1.1852384375748862e-05, | |
| "loss": 0.2906, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 1.1373112868439973e-05, | |
| "loss": 0.2792, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 1.0893841361131082e-05, | |
| "loss": 0.2777, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 1.0414569853822191e-05, | |
| "loss": 0.2786, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 9.9352983465133e-06, | |
| "loss": 0.2707, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 9.45602683920441e-06, | |
| "loss": 0.2628, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 8.976755331895519e-06, | |
| "loss": 0.2794, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 8.497483824586628e-06, | |
| "loss": 0.2591, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8.018212317277739e-06, | |
| "loss": 0.2715, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8908100178141046, | |
| "eval_f1": 0.8893163070910289, | |
| "eval_kappa": 0.5958280873418562, | |
| "eval_loss": 0.3180936276912689, | |
| "eval_precision": 0.8881406833580966, | |
| "eval_recall": 0.8908100178141046, | |
| "eval_runtime": 100.3133, | |
| "eval_samples_per_second": 95.132, | |
| "step": 12519 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 7.538940809968847e-06, | |
| "loss": 0.2559, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 7.0596693026599574e-06, | |
| "loss": 0.2426, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 6.580397795351067e-06, | |
| "loss": 0.2494, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 6.101126288042176e-06, | |
| "loss": 0.2481, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 5.621854780733286e-06, | |
| "loss": 0.2438, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 5.142583273424395e-06, | |
| "loss": 0.2599, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 4.663311766115504e-06, | |
| "loss": 0.2401, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 4.1840402588066145e-06, | |
| "loss": 0.2553, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8920674840197003, | |
| "eval_f1": 0.8862888705299119, | |
| "eval_kappa": 0.5716616874732978, | |
| "eval_loss": 0.37104833126068115, | |
| "eval_precision": 0.884886094916698, | |
| "eval_recall": 0.8920674840197003, | |
| "eval_runtime": 100.4865, | |
| "eval_samples_per_second": 94.968, | |
| "step": 16692 | |
| } | |
| ], | |
| "max_steps": 20865, | |
| "num_train_epochs": 5, | |
| "total_flos": 4.294687728722784e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |