| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 30.0, | |
| "global_step": 8280, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 2.818840579710145e-05, | |
| "loss": 1.483, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 2.63768115942029e-05, | |
| "loss": 0.5081, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "learning_rate": 2.456521739130435e-05, | |
| "loss": 0.1953, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "learning_rate": 2.2753623188405797e-05, | |
| "loss": 0.0984, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "learning_rate": 2.0942028985507246e-05, | |
| "loss": 0.0663, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 10.87, | |
| "learning_rate": 1.9130434782608694e-05, | |
| "loss": 0.0443, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 12.68, | |
| "learning_rate": 1.7318840579710146e-05, | |
| "loss": 0.0337, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 14.49, | |
| "learning_rate": 1.5507246376811594e-05, | |
| "loss": 0.0289, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 16.3, | |
| "learning_rate": 1.3695652173913042e-05, | |
| "loss": 0.0179, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 18.12, | |
| "learning_rate": 1.1884057971014492e-05, | |
| "loss": 0.0134, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 19.93, | |
| "learning_rate": 1.0072463768115942e-05, | |
| "loss": 0.0119, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 21.74, | |
| "learning_rate": 8.260869565217392e-06, | |
| "loss": 0.0066, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 23.55, | |
| "learning_rate": 6.4492753623188414e-06, | |
| "loss": 0.0095, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 25.36, | |
| "learning_rate": 4.63768115942029e-06, | |
| "loss": 0.0045, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 27.17, | |
| "learning_rate": 2.8260869565217393e-06, | |
| "loss": 0.0037, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 28.99, | |
| "learning_rate": 1.0144927536231885e-06, | |
| "loss": 0.0019, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "step": 8280, | |
| "total_flos": 2.5286585356944e+16, | |
| "train_loss": 0.1526769723581231, | |
| "train_runtime": 2830.1977, | |
| "train_samples_per_second": 87.535, | |
| "train_steps_per_second": 2.926 | |
| } | |
| ], | |
| "max_steps": 8280, | |
| "num_train_epochs": 30, | |
| "total_flos": 2.5286585356944e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |