| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.875, | |
| "eval_steps": 1, | |
| "global_step": 4, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 1.297121286392212, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 4.7213, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 2.8201005458831787, | |
| "eval_runtime": 212.0162, | |
| "eval_samples_per_second": 2.679, | |
| "eval_steps_per_second": 0.335, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.0007646083831787, | |
| "learning_rate": 0.0001, | |
| "loss": 4.4683, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.806816339492798, | |
| "eval_runtime": 222.8797, | |
| "eval_samples_per_second": 2.548, | |
| "eval_steps_per_second": 0.319, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 1.375, | |
| "grad_norm": 2.5582826137542725, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 3.6748, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 1.375, | |
| "eval_loss": 2.806488037109375, | |
| "eval_runtime": 219.2397, | |
| "eval_samples_per_second": 2.591, | |
| "eval_steps_per_second": 0.324, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 1.6264467239379883, | |
| "learning_rate": 0.0, | |
| "loss": 3.0455, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "eval_loss": 2.807661771774292, | |
| "eval_runtime": 224.703, | |
| "eval_samples_per_second": 2.528, | |
| "eval_steps_per_second": 0.316, | |
| "step": 4 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 4, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 2, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.082299413358182e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |