| { | |
| "best_metric": 0.996, | |
| "best_model_checkpoint": "xlm-roberta-base-finetuned-enron/checkpoint-7932", | |
| "epoch": 4.0, | |
| "global_step": 7932, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 1.8999495713565305e-05, | |
| "loss": 0.1679, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 1.799293998991427e-05, | |
| "loss": 0.0773, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 1.698436712052446e-05, | |
| "loss": 0.0504, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.995, | |
| "eval_loss": 0.022810379043221474, | |
| "eval_runtime": 7.2177, | |
| "eval_samples_per_second": 277.095, | |
| "eval_steps_per_second": 17.318, | |
| "step": 1983 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 1.5977811396873426e-05, | |
| "loss": 0.0436, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 1.4969238527483611e-05, | |
| "loss": 0.0377, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 1.39606656580938e-05, | |
| "loss": 0.0283, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 1.2952092788703986e-05, | |
| "loss": 0.0211, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.991, | |
| "eval_loss": 0.05412155017256737, | |
| "eval_runtime": 8.3181, | |
| "eval_samples_per_second": 240.441, | |
| "eval_steps_per_second": 15.028, | |
| "step": 3966 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 1.194351991931417e-05, | |
| "loss": 0.0201, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 1.0936964195663137e-05, | |
| "loss": 0.0073, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 9.928391326273324e-06, | |
| "loss": 0.0094, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 8.919818456883512e-06, | |
| "loss": 0.0118, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.993, | |
| "eval_loss": 0.0558784045279026, | |
| "eval_runtime": 7.2776, | |
| "eval_samples_per_second": 274.815, | |
| "eval_steps_per_second": 17.176, | |
| "step": 5949 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 7.911245587493697e-06, | |
| "loss": 0.0122, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 6.902672718103883e-06, | |
| "loss": 0.0034, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 5.89409984871407e-06, | |
| "loss": 0.0091, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 4.885526979324257e-06, | |
| "loss": 0.0027, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.996, | |
| "eval_loss": 0.031464751809835434, | |
| "eval_runtime": 7.3051, | |
| "eval_samples_per_second": 273.782, | |
| "eval_steps_per_second": 17.111, | |
| "step": 7932 | |
| } | |
| ], | |
| "max_steps": 9915, | |
| "num_train_epochs": 5, | |
| "total_flos": 3.326409472938432e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |