| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "global_step": 8780, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 4.715261958997722e-05, | |
| "loss": 0.3945, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.4305239179954444e-05, | |
| "loss": 0.1883, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 4.1457858769931665e-05, | |
| "loss": 0.1584, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 3.8610478359908886e-05, | |
| "loss": 0.1311, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 3.5763097949886106e-05, | |
| "loss": 0.1052, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 3.291571753986333e-05, | |
| "loss": 0.0989, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 3.0068337129840545e-05, | |
| "loss": 0.0868, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 2.722095671981777e-05, | |
| "loss": 0.0687, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 2.437357630979499e-05, | |
| "loss": 0.0694, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 2.152619589977221e-05, | |
| "loss": 0.0594, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 1.867881548974943e-05, | |
| "loss": 0.0556, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 1.5831435079726652e-05, | |
| "loss": 0.0498, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 1.2984054669703875e-05, | |
| "loss": 0.0472, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 1.0136674259681094e-05, | |
| "loss": 0.0439, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 7.289293849658315e-06, | |
| "loss": 0.0373, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "learning_rate": 4.4419134396355355e-06, | |
| "loss": 0.0369, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 1.5945330296127566e-06, | |
| "loss": 0.0369, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 8780, | |
| "total_flos": 55398132371610.0, | |
| "train_loss": 0.09602124729026151, | |
| "train_runtime": 167.8929, | |
| "train_samples_per_second": 418.153, | |
| "train_steps_per_second": 52.295 | |
| } | |
| ], | |
| "max_steps": 8780, | |
| "num_train_epochs": 5, | |
| "total_flos": 55398132371610.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |