| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.00843775049571784, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00016875500991435683, | |
| "eval_loss": 1.6930816173553467, | |
| "eval_runtime": 57.8268, | |
| "eval_samples_per_second": 43.146, | |
| "eval_steps_per_second": 21.582, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0016875500991435683, | |
| "grad_norm": 3.9818174839019775, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 1.7126, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0021938151288866388, | |
| "eval_loss": 1.502448320388794, | |
| "eval_runtime": 56.5446, | |
| "eval_samples_per_second": 44.124, | |
| "eval_steps_per_second": 22.071, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0033751001982871365, | |
| "grad_norm": 5.268524169921875, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 1.5099, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0043876302577732776, | |
| "eval_loss": 1.3147406578063965, | |
| "eval_runtime": 56.6894, | |
| "eval_samples_per_second": 44.012, | |
| "eval_steps_per_second": 22.015, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.005062650297430705, | |
| "grad_norm": 8.063873291015625, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 1.3195, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.006581445386659917, | |
| "eval_loss": 1.2810492515563965, | |
| "eval_runtime": 56.5925, | |
| "eval_samples_per_second": 44.087, | |
| "eval_steps_per_second": 22.052, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.006750200396574273, | |
| "grad_norm": 6.396270751953125, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 1.3132, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.00843775049571784, | |
| "grad_norm": 6.293328285217285, | |
| "learning_rate": 0.0, | |
| "loss": 1.2793, | |
| "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": 3269109743616000.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |