| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.01253616200578592, | |
| "eval_steps": 10, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0003214400514304082, | |
| "eval_loss": 2.0185434818267822, | |
| "eval_runtime": 146.345, | |
| "eval_samples_per_second": 8.951, | |
| "eval_steps_per_second": 4.476, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0016072002571520412, | |
| "grad_norm": 2.316012382507324, | |
| "learning_rate": 5e-05, | |
| "loss": 8.2139, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0032144005143040825, | |
| "grad_norm": 2.698329448699951, | |
| "learning_rate": 0.0001, | |
| "loss": 7.9446, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0032144005143040825, | |
| "eval_loss": 1.9448484182357788, | |
| "eval_runtime": 147.01, | |
| "eval_samples_per_second": 8.911, | |
| "eval_steps_per_second": 4.455, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0048216007714561235, | |
| "grad_norm": 2.331204414367676, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 7.4394, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.006428801028608165, | |
| "grad_norm": 3.3225786685943604, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 7.0552, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.006428801028608165, | |
| "eval_loss": 1.8020565509796143, | |
| "eval_runtime": 147.1652, | |
| "eval_samples_per_second": 8.902, | |
| "eval_steps_per_second": 4.451, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.008036001285760206, | |
| "grad_norm": 4.083289623260498, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 6.9482, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.009643201542912247, | |
| "grad_norm": 2.6887266635894775, | |
| "learning_rate": 5e-05, | |
| "loss": 6.9021, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.009643201542912247, | |
| "eval_loss": 1.7518081665039062, | |
| "eval_runtime": 147.1029, | |
| "eval_samples_per_second": 8.905, | |
| "eval_steps_per_second": 4.453, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.011250401800064288, | |
| "grad_norm": 2.600066661834717, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 7.3319, | |
| "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": 1.540514893529088e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |