| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 9117, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 4.725786991334869e-05, | |
| "loss": 3.0811, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.451573982669738e-05, | |
| "loss": 2.9144, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.177360974004607e-05, | |
| "loss": 2.861, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 3.903147965339476e-05, | |
| "loss": 2.8143, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 3.628934956674345e-05, | |
| "loss": 2.7832, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 3.354721948009214e-05, | |
| "loss": 2.7602, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 3.0805089393440825e-05, | |
| "loss": 2.7305, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 2.8062959306789516e-05, | |
| "loss": 2.7235, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 2.5320829220138204e-05, | |
| "loss": 2.7046, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 2.2578699133486895e-05, | |
| "loss": 2.6897, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 1.9836569046835583e-05, | |
| "loss": 2.6826, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 1.709443896018427e-05, | |
| "loss": 2.667, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 1.4352308873532961e-05, | |
| "loss": 2.6481, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 1.161017878688165e-05, | |
| "loss": 2.647, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 8.86804870023034e-06, | |
| "loss": 2.644, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 6.125918613579029e-06, | |
| "loss": 2.6336, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 3.3837885269277177e-06, | |
| "loss": 2.6286, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 6.416584402764068e-07, | |
| "loss": 2.6227, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 9117, | |
| "total_flos": 8.259990898448794e+16, | |
| "train_loss": 2.733830861908257, | |
| "train_runtime": 14258.7293, | |
| "train_samples_per_second": 20.458, | |
| "train_steps_per_second": 0.639 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 9117, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 8.259990898448794e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |