| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0022326699727356646, | |
| "eval_steps": 10, | |
| "global_step": 26, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 8.58719220282948e-05, | |
| "eval_loss": 3.8520586490631104, | |
| "eval_runtime": 1290.9347, | |
| "eval_samples_per_second": 3.799, | |
| "eval_steps_per_second": 1.899, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.000429359610141474, | |
| "grad_norm": 13.166454315185547, | |
| "learning_rate": 5e-05, | |
| "loss": 12.8444, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.000858719220282948, | |
| "grad_norm": 16.13837432861328, | |
| "learning_rate": 0.0001, | |
| "loss": 14.1954, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.000858719220282948, | |
| "eval_loss": 3.2008979320526123, | |
| "eval_runtime": 3808.0538, | |
| "eval_samples_per_second": 1.288, | |
| "eval_steps_per_second": 0.644, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.001288078830424422, | |
| "grad_norm": 13.235127449035645, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 9.7988, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.001717438440565896, | |
| "grad_norm": 20.243595123291016, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 6.6174, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.001717438440565896, | |
| "eval_loss": 1.4080783128738403, | |
| "eval_runtime": 3229.1069, | |
| "eval_samples_per_second": 1.519, | |
| "eval_steps_per_second": 0.759, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.00214679805070737, | |
| "grad_norm": 18.120956420898438, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 5.2192, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4519734452158464.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |