| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.076923076923077, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 0.6670007587829014, | |
| "learning_rate": 0.0001, | |
| "loss": 1.0372, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 0.8347824338177103, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 0.9638, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "eval_loss": 0.7346013784408569, | |
| "eval_runtime": 2.5552, | |
| "eval_samples_per_second": 7.827, | |
| "eval_steps_per_second": 1.174, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 0.3344819405790827, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 0.6247, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "eval_loss": 0.6349390149116516, | |
| "eval_runtime": 1.3276, | |
| "eval_samples_per_second": 15.065, | |
| "eval_steps_per_second": 2.26, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 0.3866613912988811, | |
| "learning_rate": 3.5721239031346066e-05, | |
| "loss": 0.5318, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "eval_loss": 0.6109390258789062, | |
| "eval_runtime": 1.3273, | |
| "eval_samples_per_second": 15.068, | |
| "eval_steps_per_second": 2.26, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 0.3445137999539839, | |
| "learning_rate": 0.0, | |
| "loss": 0.4704, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "eval_loss": 0.6075538396835327, | |
| "eval_runtime": 1.3265, | |
| "eval_samples_per_second": 15.078, | |
| "eval_steps_per_second": 2.262, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "step": 20, | |
| "total_flos": 2502654197760.0, | |
| "train_loss": 0.6513590574264526, | |
| "train_runtime": 83.5417, | |
| "train_samples_per_second": 3.83, | |
| "train_steps_per_second": 0.239 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2502654197760.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |