| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.05739514348785872, | |
| "eval_steps": 10, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0014716703458425313, | |
| "eval_loss": 1.1566596031188965, | |
| "eval_runtime": 26.6038, | |
| "eval_samples_per_second": 10.788, | |
| "eval_steps_per_second": 5.413, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.007358351729212656, | |
| "grad_norm": 16.61760139465332, | |
| "learning_rate": 5e-05, | |
| "loss": 3.9117, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.014716703458425313, | |
| "grad_norm": 6.252457618713379, | |
| "learning_rate": 0.0001, | |
| "loss": 3.6101, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.014716703458425313, | |
| "eval_loss": 0.9047649502754211, | |
| "eval_runtime": 27.0546, | |
| "eval_samples_per_second": 10.608, | |
| "eval_steps_per_second": 5.323, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02207505518763797, | |
| "grad_norm": 10.30296516418457, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 3.2339, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.029433406916850625, | |
| "grad_norm": 6.257486343383789, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 4.0174, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.029433406916850625, | |
| "eval_loss": 0.865932047367096, | |
| "eval_runtime": 27.0244, | |
| "eval_samples_per_second": 10.62, | |
| "eval_steps_per_second": 5.329, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03679175864606328, | |
| "grad_norm": 7.642504692077637, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 3.1607, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.04415011037527594, | |
| "grad_norm": 11.923357963562012, | |
| "learning_rate": 5e-05, | |
| "loss": 3.7893, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04415011037527594, | |
| "eval_loss": 0.8530696630477905, | |
| "eval_runtime": 27.2736, | |
| "eval_samples_per_second": 10.523, | |
| "eval_steps_per_second": 5.28, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.051508462104488596, | |
| "grad_norm": 6.583628177642822, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 3.433, | |
| "step": 35 | |
| } | |
| ], | |
| "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": 1.0345312992362496e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |