| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9473684210526315, | |
| "eval_steps": 9, | |
| "global_step": 36, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02631578947368421, | |
| "eval_loss": 1.6184024810791016, | |
| "eval_runtime": 4.4294, | |
| "eval_samples_per_second": 14.449, | |
| "eval_steps_per_second": 1.806, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07894736842105263, | |
| "grad_norm": 26.1455020904541, | |
| "learning_rate": 3e-05, | |
| "loss": 6.2533, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.15789473684210525, | |
| "grad_norm": 20.71628761291504, | |
| "learning_rate": 6e-05, | |
| "loss": 6.0813, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.23684210526315788, | |
| "grad_norm": 17.294282913208008, | |
| "learning_rate": 9e-05, | |
| "loss": 4.9981, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.23684210526315788, | |
| "eval_loss": 1.035604476928711, | |
| "eval_runtime": 4.5385, | |
| "eval_samples_per_second": 14.102, | |
| "eval_steps_per_second": 1.763, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.3157894736842105, | |
| "grad_norm": 11.629775047302246, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 4.0086, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.39473684210526316, | |
| "grad_norm": 10.649725914001465, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 3.8607, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.47368421052631576, | |
| "grad_norm": 9.19792366027832, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 3.954, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.47368421052631576, | |
| "eval_loss": 0.8341699838638306, | |
| "eval_runtime": 4.56, | |
| "eval_samples_per_second": 14.035, | |
| "eval_steps_per_second": 1.754, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.5526315789473685, | |
| "grad_norm": 8.501199722290039, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 3.2274, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.631578947368421, | |
| "grad_norm": 10.514701843261719, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 3.3657, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.7105263157894737, | |
| "grad_norm": 8.145644187927246, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 3.5328, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.7105263157894737, | |
| "eval_loss": 0.7960277199745178, | |
| "eval_runtime": 4.5667, | |
| "eval_samples_per_second": 14.015, | |
| "eval_steps_per_second": 1.752, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.7894736842105263, | |
| "grad_norm": 9.123873710632324, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 3.727, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.868421052631579, | |
| "grad_norm": 9.119816780090332, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 3.2364, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.9473684210526315, | |
| "grad_norm": 9.756755828857422, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 3.4796, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.9473684210526315, | |
| "eval_loss": 0.7768706679344177, | |
| "eval_runtime": 4.5592, | |
| "eval_samples_per_second": 14.038, | |
| "eval_steps_per_second": 1.755, | |
| "step": 36 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.062538188711526e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |