| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0576414595452142, | |
| "eval_steps": 1000, | |
| "global_step": 4000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13220518244315177, | |
| "grad_norm": 0.8546391725540161, | |
| "learning_rate": 8.816009873931059e-05, | |
| "loss": 5.1118, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.26441036488630354, | |
| "grad_norm": 0.8593688607215881, | |
| "learning_rate": 9.59831475011252e-05, | |
| "loss": 3.406, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.26441036488630354, | |
| "eval_accuracy": 0.5035306174465283, | |
| "eval_loss": 3.23445987701416, | |
| "eval_runtime": 73.8676, | |
| "eval_samples_per_second": 24.909, | |
| "eval_steps_per_second": 0.528, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.3966155473294553, | |
| "grad_norm": 0.9617258906364441, | |
| "learning_rate": 9.134314230431938e-05, | |
| "loss": 3.0005, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.5288207297726071, | |
| "grad_norm": 0.8953185677528381, | |
| "learning_rate": 8.670313710751356e-05, | |
| "loss": 2.8119, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.5288207297726071, | |
| "eval_accuracy": 0.5365118094348038, | |
| "eval_loss": 2.821384906768799, | |
| "eval_runtime": 72.909, | |
| "eval_samples_per_second": 25.237, | |
| "eval_steps_per_second": 0.535, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.6610259122157589, | |
| "grad_norm": 1.4154396057128906, | |
| "learning_rate": 8.206313191070773e-05, | |
| "loss": 2.686, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.7932310946589106, | |
| "grad_norm": 1.821349024772644, | |
| "learning_rate": 7.742312671390191e-05, | |
| "loss": 2.607, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.7932310946589106, | |
| "eval_accuracy": 0.5497897240925214, | |
| "eval_loss": 2.657219886779785, | |
| "eval_runtime": 73.4297, | |
| "eval_samples_per_second": 25.058, | |
| "eval_steps_per_second": 0.531, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.9254362771020624, | |
| "grad_norm": 2.0297396183013916, | |
| "learning_rate": 7.278312151709609e-05, | |
| "loss": 2.5642, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.0576414595452142, | |
| "grad_norm": 2.8318285942077637, | |
| "learning_rate": 6.814311632029027e-05, | |
| "loss": 2.4734, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.0576414595452142, | |
| "eval_accuracy": 0.5582058048894458, | |
| "eval_loss": 2.5735702514648438, | |
| "eval_runtime": 73.4679, | |
| "eval_samples_per_second": 25.045, | |
| "eval_steps_per_second": 0.531, | |
| "step": 4000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 11346, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.124838835670221e+16, | |
| "train_batch_size": 24, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |