| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9900990099009901, | |
| "eval_steps": 10, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.19801980198019803, | |
| "grad_norm": 180.49745178222656, | |
| "learning_rate": 0.00017600000000000002, | |
| "loss": 6.1581, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.39603960396039606, | |
| "grad_norm": 12.116339683532715, | |
| "learning_rate": 0.00013600000000000003, | |
| "loss": 0.5783, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.39603960396039606, | |
| "eval_loss": 0.2537325918674469, | |
| "eval_runtime": 48.0526, | |
| "eval_samples_per_second": 1.686, | |
| "eval_steps_per_second": 0.853, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.594059405940594, | |
| "grad_norm": 3.929807186126709, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.2015, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.7920792079207921, | |
| "grad_norm": 2.739349842071533, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 0.1547, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7920792079207921, | |
| "eval_loss": 0.1523071825504303, | |
| "eval_runtime": 47.9248, | |
| "eval_samples_per_second": 1.69, | |
| "eval_steps_per_second": 0.856, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.9900990099009901, | |
| "grad_norm": 1.2208682298660278, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.1366, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 25, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1759069725802752.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |