| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.020766275568476794, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0004153255113695359, | |
| "eval_loss": 0.9820240139961243, | |
| "eval_runtime": 57.8517, | |
| "eval_samples_per_second": 17.528, | |
| "eval_steps_per_second": 8.764, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0041532551136953585, | |
| "grad_norm": 0.565779447555542, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 0.8725, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.005399231647803967, | |
| "eval_loss": 0.6550027132034302, | |
| "eval_runtime": 57.3561, | |
| "eval_samples_per_second": 17.679, | |
| "eval_steps_per_second": 8.84, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.008306510227390717, | |
| "grad_norm": 0.30139827728271484, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.652, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.010798463295607933, | |
| "eval_loss": 0.587783694267273, | |
| "eval_runtime": 57.3749, | |
| "eval_samples_per_second": 17.673, | |
| "eval_steps_per_second": 8.837, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.012459765341086075, | |
| "grad_norm": 0.3119395077228546, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 0.6132, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0161976949434119, | |
| "eval_loss": 0.5670517086982727, | |
| "eval_runtime": 57.3936, | |
| "eval_samples_per_second": 17.667, | |
| "eval_steps_per_second": 8.834, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.016613020454781434, | |
| "grad_norm": 0.26189422607421875, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 0.5679, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.020766275568476794, | |
| "grad_norm": 0.269429087638855, | |
| "learning_rate": 0.0, | |
| "loss": 0.5731, | |
| "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": 1.6478229233664e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |