| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.002739444577611889, | |
| "eval_steps": 3, | |
| "global_step": 10, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0002739444577611889, | |
| "grad_norm": NaN, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0002739444577611889, | |
| "eval_loss": NaN, | |
| "eval_runtime": 289.6935, | |
| "eval_samples_per_second": 21.222, | |
| "eval_steps_per_second": 10.611, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0005478889155223778, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0008218333732835667, | |
| "grad_norm": NaN, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0008218333732835667, | |
| "eval_loss": NaN, | |
| "eval_runtime": 164.6021, | |
| "eval_samples_per_second": 37.351, | |
| "eval_steps_per_second": 18.675, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0010957778310447556, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0013697222888059445, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0016436667465671334, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0016436667465671334, | |
| "eval_loss": NaN, | |
| "eval_runtime": 164.5567, | |
| "eval_samples_per_second": 37.361, | |
| "eval_steps_per_second": 18.68, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0019176112043283223, | |
| "grad_norm": NaN, | |
| "learning_rate": 7e-05, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0021915556620895112, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0024655001198507004, | |
| "grad_norm": NaN, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0024655001198507004, | |
| "eval_loss": NaN, | |
| "eval_runtime": 164.9532, | |
| "eval_samples_per_second": 37.271, | |
| "eval_steps_per_second": 18.636, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.002739444577611889, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 10 | |
| } | |
| ], | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 873631076843520.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |