| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.02844950213371266, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0005689900426742532, | |
| "eval_loss": 1.6969337463378906, | |
| "eval_runtime": 25.7196, | |
| "eval_samples_per_second": 28.772, | |
| "eval_steps_per_second": 14.386, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.005689900426742532, | |
| "grad_norm": 3.72646427154541, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 6.4051, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.007396870554765292, | |
| "eval_loss": 1.508939504623413, | |
| "eval_runtime": 25.5041, | |
| "eval_samples_per_second": 29.015, | |
| "eval_steps_per_second": 14.507, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.011379800853485065, | |
| "grad_norm": 5.646580219268799, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 6.0452, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.014793741109530583, | |
| "eval_loss": 1.4555972814559937, | |
| "eval_runtime": 25.7786, | |
| "eval_samples_per_second": 28.706, | |
| "eval_steps_per_second": 14.353, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.017069701280227598, | |
| "grad_norm": 12.445784568786621, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 5.6335, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.022190611664295874, | |
| "eval_loss": 1.4427510499954224, | |
| "eval_runtime": 25.7036, | |
| "eval_samples_per_second": 28.79, | |
| "eval_steps_per_second": 14.395, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.02275960170697013, | |
| "grad_norm": 5.738654136657715, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 5.6701, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.02844950213371266, | |
| "grad_norm": 5.592751979827881, | |
| "learning_rate": 0.0, | |
| "loss": 5.0778, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8591567291940864.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |