| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 50, | |
| "global_step": 4338, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11526048870447211, | |
| "grad_norm": 20.16172218322754, | |
| "learning_rate": 9.960000000000001e-06, | |
| "loss": 2.3764, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.23052097740894423, | |
| "grad_norm": 16.672468185424805, | |
| "learning_rate": 8.705054715997916e-06, | |
| "loss": 0.5451, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.3457814661134163, | |
| "grad_norm": 11.151906967163086, | |
| "learning_rate": 7.402292860865034e-06, | |
| "loss": 0.4302, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.46104195481788846, | |
| "grad_norm": 7.392632961273193, | |
| "learning_rate": 6.099531005732152e-06, | |
| "loss": 0.3772, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.5763024435223605, | |
| "grad_norm": 6.710408687591553, | |
| "learning_rate": 4.796769150599271e-06, | |
| "loss": 0.3742, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.6915629322268326, | |
| "grad_norm": 8.172633171081543, | |
| "learning_rate": 3.494007295466389e-06, | |
| "loss": 0.3518, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.8068234209313048, | |
| "grad_norm": 5.664760112762451, | |
| "learning_rate": 2.1912454403335073e-06, | |
| "loss": 0.3619, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.9220839096357769, | |
| "grad_norm": 16.45347785949707, | |
| "learning_rate": 8.884835852006255e-07, | |
| "loss": 0.3293, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 4338, | |
| "total_flos": 4.801972373497037e+16, | |
| "train_loss": 0.6182623529500156, | |
| "train_runtime": 1876.03, | |
| "train_samples_per_second": 36.992, | |
| "train_steps_per_second": 2.312 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4338, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.801972373497037e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |