| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 1479, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 2e-05, | |
| "loss": 3.6557, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 2e-05, | |
| "loss": 1.6581, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 2e-05, | |
| "loss": 1.5014, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 2e-05, | |
| "loss": 1.3948, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.1438196897506714, | |
| "eval_macro_f1": 0.43287074814809323, | |
| "eval_runtime": 191.8217, | |
| "eval_samples_per_second": 7.059, | |
| "eval_steps_per_second": 1.413, | |
| "eval_w_f1": 0.5734991666701943, | |
| "step": 493 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 2e-05, | |
| "loss": 1.2719, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 2e-05, | |
| "loss": 1.1415, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0834, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 2e-05, | |
| "loss": 1.1151, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 2e-05, | |
| "loss": 1.1067, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.6219320297241211, | |
| "eval_macro_f1": 0.698736427313761, | |
| "eval_runtime": 192.5646, | |
| "eval_samples_per_second": 7.031, | |
| "eval_steps_per_second": 1.407, | |
| "eval_w_f1": 0.7795811904122447, | |
| "step": 986 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0408, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5776, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6012, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6464, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6282, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.21291832625865936, | |
| "eval_macro_f1": 0.919558189524972, | |
| "eval_runtime": 190.235, | |
| "eval_samples_per_second": 7.118, | |
| "eval_steps_per_second": 1.425, | |
| "eval_w_f1": 0.946814493022281, | |
| "step": 1479 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 2465, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 3.163537548829901e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |