| { |
| "best_global_step": 1524, |
| "best_metric": 0.8752606955526183, |
| "best_model_checkpoint": "../models/checkpoint-1524", |
| "epoch": 3.0, |
| "eval_steps": 500, |
| "global_step": 1524, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 6.651031970977783, |
| "learning_rate": 1.3346456692913388e-05, |
| "loss": 0.4228, |
| "step": 508 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_f1_macro": 0.843772475820936, |
| "eval_f1_micro": 0.8440423247968103, |
| "eval_loss": 0.30257782340049744, |
| "eval_precision": 0.810126582278481, |
| "eval_recall": 0.8809218950064021, |
| "eval_runtime": 15.8588, |
| "eval_samples_per_second": 256.072, |
| "eval_steps_per_second": 4.036, |
| "step": 508 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 7.698607444763184, |
| "learning_rate": 6.6797900262467195e-06, |
| "loss": 0.2626, |
| "step": 1016 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_f1_macro": 0.8678396468883778, |
| "eval_f1_micro": 0.8675043741052967, |
| "eval_loss": 0.2636893689632416, |
| "eval_precision": 0.8621561808409738, |
| "eval_recall": 0.8729193341869398, |
| "eval_runtime": 15.5885, |
| "eval_samples_per_second": 260.513, |
| "eval_steps_per_second": 4.106, |
| "step": 1016 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 13.880615234375, |
| "learning_rate": 1.3123359580052495e-08, |
| "loss": 0.2073, |
| "step": 1524 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_f1_macro": 0.8752606955526183, |
| "eval_f1_micro": 0.8743307086614174, |
| "eval_loss": 0.2604271173477173, |
| "eval_precision": 0.8605083694978302, |
| "eval_recall": 0.8886043533930857, |
| "eval_runtime": 15.6076, |
| "eval_samples_per_second": 260.194, |
| "eval_steps_per_second": 4.101, |
| "step": 1524 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 1524, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.282113861663744e+16, |
| "train_batch_size": 32, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
