| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9997758637251448, | |
| "global_step": 7806, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.6797335383038696e-05, | |
| "loss": 1.7853, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.3594670766077376e-05, | |
| "loss": 1.3491, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 4.039200614911607e-05, | |
| "loss": 1.2577, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 3.7189341532154756e-05, | |
| "loss": 1.2039, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 3.398667691519344e-05, | |
| "loss": 1.1695, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 3.078401229823213e-05, | |
| "loss": 1.1321, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 2.758134768127082e-05, | |
| "loss": 1.1055, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 2.4378683064309507e-05, | |
| "loss": 1.0857, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 2.1176018447348194e-05, | |
| "loss": 1.0507, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 1.7973353830386884e-05, | |
| "loss": 1.0375, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 1.4770689213425571e-05, | |
| "loss": 1.0249, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 1.156802459646426e-05, | |
| "loss": 1.0106, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 8.365359979502948e-06, | |
| "loss": 1.0087, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 5.162695362541635e-06, | |
| "loss": 1.0034, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 1.960030745580323e-06, | |
| "loss": 0.997, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 7806, | |
| "total_flos": 1.5713593290748068e+18, | |
| "train_loss": 1.142376601864001, | |
| "train_runtime": 22889.3512, | |
| "train_samples_per_second": 130.985, | |
| "train_steps_per_second": 0.341 | |
| } | |
| ], | |
| "max_steps": 7806, | |
| "num_train_epochs": 2, | |
| "total_flos": 1.5713593290748068e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |