| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.02126528442317916, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0004253056884635832, | |
| "eval_loss": 2.1829206943511963, | |
| "eval_runtime": 52.1069, | |
| "eval_samples_per_second": 18.999, | |
| "eval_steps_per_second": 9.5, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004253056884635832, | |
| "grad_norm": 3.2505462169647217, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 7.9491, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.005528973950026581, | |
| "eval_loss": 1.6017568111419678, | |
| "eval_runtime": 52.0996, | |
| "eval_samples_per_second": 19.002, | |
| "eval_steps_per_second": 9.501, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.008506113769271665, | |
| "grad_norm": 2.187023639678955, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 6.2868, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.011057947900053162, | |
| "eval_loss": 1.4111206531524658, | |
| "eval_runtime": 52.0517, | |
| "eval_samples_per_second": 19.02, | |
| "eval_steps_per_second": 9.51, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.012759170653907496, | |
| "grad_norm": 2.315171480178833, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 5.5634, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.016586921850079744, | |
| "eval_loss": 1.3345240354537964, | |
| "eval_runtime": 51.9662, | |
| "eval_samples_per_second": 19.051, | |
| "eval_steps_per_second": 9.525, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.01701222753854333, | |
| "grad_norm": 2.433863401412964, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 5.7079, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.02126528442317916, | |
| "grad_norm": 2.022089719772339, | |
| "learning_rate": 0.0, | |
| "loss": 5.2807, | |
| "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": 1.1811201302839296e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |