| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0007027304004099261, | |
| "eval_steps": 3, | |
| "global_step": 9, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 7.80811556011029e-05, | |
| "grad_norm": NaN, | |
| "learning_rate": 1e-05, | |
| "loss": 43349.4531, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 7.80811556011029e-05, | |
| "eval_loss": NaN, | |
| "eval_runtime": 783.7756, | |
| "eval_samples_per_second": 27.521, | |
| "eval_steps_per_second": 13.76, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0001561623112022058, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00023424346680330868, | |
| "grad_norm": NaN, | |
| "learning_rate": 3e-05, | |
| "loss": 5.8715, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00023424346680330868, | |
| "eval_loss": NaN, | |
| "eval_runtime": 697.1851, | |
| "eval_samples_per_second": 30.939, | |
| "eval_steps_per_second": 15.469, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0003123246224044116, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0003904057780055145, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00046848693360661736, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 34956.0312, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00046848693360661736, | |
| "eval_loss": NaN, | |
| "eval_runtime": 697.3582, | |
| "eval_samples_per_second": 30.931, | |
| "eval_steps_per_second": 15.466, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0005465680892077203, | |
| "grad_norm": NaN, | |
| "learning_rate": 7e-05, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0006246492448088232, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 253.8839, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0007027304004099261, | |
| "grad_norm": NaN, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0007027304004099261, | |
| "eval_loss": NaN, | |
| "eval_runtime": 697.6516, | |
| "eval_samples_per_second": 30.918, | |
| "eval_steps_per_second": 15.459, | |
| "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": 9552061305716736.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |