| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0008531210009953079, | |
| "eval_steps": 3, | |
| "global_step": 9, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 9.479122233281199e-05, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 9.479122233281199e-05, | |
| "eval_loss": NaN, | |
| "eval_runtime": 249.6567, | |
| "eval_samples_per_second": 17.792, | |
| "eval_steps_per_second": 8.896, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00018958244466562397, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00028437366699843596, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00028437366699843596, | |
| "eval_loss": NaN, | |
| "eval_runtime": 248.1124, | |
| "eval_samples_per_second": 17.903, | |
| "eval_steps_per_second": 8.952, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00037916488933124794, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00047395611166405993, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0005687473339968719, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0005687473339968719, | |
| "eval_loss": NaN, | |
| "eval_runtime": 248.0649, | |
| "eval_samples_per_second": 17.907, | |
| "eval_steps_per_second": 8.953, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0006635385563296839, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00014, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0007583297786624959, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0008531210009953079, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0008531210009953079, | |
| "eval_loss": NaN, | |
| "eval_runtime": 248.1797, | |
| "eval_samples_per_second": 17.898, | |
| "eval_steps_per_second": 8.949, | |
| "step": 9 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 10, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 3, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3135477657894912.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |