| { | |
| "best_metric": 18.7407, | |
| "best_model_checkpoint": "saved/tobyoki-pairwise-wo_space/bart-base-japanese/BaseModel/checkpoint-10125", | |
| "epoch": 5.0, | |
| "global_step": 10125, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.7530864197530866e-05, | |
| "loss": 2.1397, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.506172839506173e-05, | |
| "loss": 1.3468, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 4.259259259259259e-05, | |
| "loss": 0.9786, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 4.012345679012346e-05, | |
| "loss": 0.7833, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_gen_len": 65.37762237762237, | |
| "eval_loss": 2.5750701427459717, | |
| "eval_rouge1": 16.3343, | |
| "eval_rouge2": 1.2888, | |
| "eval_rougeL": 11.0128, | |
| "eval_rougeLsum": 15.2802, | |
| "eval_runtime": 267.3657, | |
| "eval_samples_per_second": 1.07, | |
| "eval_steps_per_second": 1.07, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 3.7654320987654326e-05, | |
| "loss": 0.5998, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 3.518518518518519e-05, | |
| "loss": 0.467, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 3.271604938271605e-05, | |
| "loss": 0.4008, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 3.0246913580246916e-05, | |
| "loss": 0.3308, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_gen_len": 91.65034965034965, | |
| "eval_loss": 2.9422976970672607, | |
| "eval_rouge1": 17.9514, | |
| "eval_rouge2": 2.8091, | |
| "eval_rougeL": 10.9133, | |
| "eval_rougeLsum": 15.4068, | |
| "eval_runtime": 356.3388, | |
| "eval_samples_per_second": 0.803, | |
| "eval_steps_per_second": 0.803, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.2778, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 2.5308641975308646e-05, | |
| "loss": 0.2482, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 2.2839506172839506e-05, | |
| "loss": 0.2349, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 2.037037037037037e-05, | |
| "loss": 0.2302, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_gen_len": 77.1993006993007, | |
| "eval_loss": 3.0624501705169678, | |
| "eval_rouge1": 16.1453, | |
| "eval_rouge2": 3.0026, | |
| "eval_rougeL": 10.272, | |
| "eval_rougeLsum": 14.0716, | |
| "eval_runtime": 279.7235, | |
| "eval_samples_per_second": 1.022, | |
| "eval_steps_per_second": 1.022, | |
| "step": 6075 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 1.7901234567901236e-05, | |
| "loss": 0.1927, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 1.54320987654321e-05, | |
| "loss": 0.1607, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 1.2962962962962962e-05, | |
| "loss": 0.1621, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 1.0493827160493827e-05, | |
| "loss": 0.1576, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_gen_len": 88.3986013986014, | |
| "eval_loss": 3.2307794094085693, | |
| "eval_rouge1": 17.8409, | |
| "eval_rouge2": 2.9937, | |
| "eval_rougeL": 10.8765, | |
| "eval_rougeLsum": 15.6203, | |
| "eval_runtime": 323.891, | |
| "eval_samples_per_second": 0.883, | |
| "eval_steps_per_second": 0.883, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "learning_rate": 8.02469135802469e-06, | |
| "loss": 0.1289, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.1385, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "learning_rate": 3.0864197530864196e-06, | |
| "loss": 0.1315, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "learning_rate": 6.17283950617284e-07, | |
| "loss": 0.1055, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_gen_len": 95.02447552447552, | |
| "eval_loss": 3.2694220542907715, | |
| "eval_rouge1": 18.7407, | |
| "eval_rouge2": 3.1211, | |
| "eval_rougeL": 10.9379, | |
| "eval_rougeLsum": 15.8203, | |
| "eval_runtime": 317.0639, | |
| "eval_samples_per_second": 0.902, | |
| "eval_steps_per_second": 0.902, | |
| "step": 10125 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 10125, | |
| "total_flos": 1931007009669120.0, | |
| "train_loss": 0.4568321267587167, | |
| "train_runtime": 2611.1815, | |
| "train_samples_per_second": 3.878, | |
| "train_steps_per_second": 3.878 | |
| } | |
| ], | |
| "max_steps": 10125, | |
| "num_train_epochs": 5, | |
| "total_flos": 1931007009669120.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |