| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3000.0, | |
| "eval_steps": 500, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 100.0, | |
| "grad_norm": 0.01587696745991707, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2798, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 200.0, | |
| "grad_norm": 0.02660752832889557, | |
| "learning_rate": 4.999688473794144e-05, | |
| "loss": 0.1875, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 300.0, | |
| "grad_norm": 0.010020197369158268, | |
| "learning_rate": 4.998753972815435e-05, | |
| "loss": 0.0391, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 400.0, | |
| "grad_norm": 0.00830842088907957, | |
| "learning_rate": 4.997196729961109e-05, | |
| "loss": 0.0036, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 500.0, | |
| "grad_norm": 0.006963910534977913, | |
| "learning_rate": 4.9950171333287335e-05, | |
| "loss": 0.0018, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 600.0, | |
| "grad_norm": 0.0036349259316921234, | |
| "learning_rate": 4.992215726119484e-05, | |
| "loss": 0.0013, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 700.0, | |
| "grad_norm": 0.008688577450811863, | |
| "learning_rate": 4.9887932065027656e-05, | |
| "loss": 0.0011, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 800.0, | |
| "grad_norm": 0.003335146466270089, | |
| "learning_rate": 4.98475042744222e-05, | |
| "loss": 0.0009, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 900.0, | |
| "grad_norm": 0.002029621507972479, | |
| "learning_rate": 4.9800883964831445e-05, | |
| "loss": 0.0009, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1000.0, | |
| "grad_norm": 0.006406690459698439, | |
| "learning_rate": 4.9748082755013934e-05, | |
| "loss": 0.0008, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1100.0, | |
| "grad_norm": 0.005226912442594767, | |
| "learning_rate": 4.96891138041381e-05, | |
| "loss": 0.0008, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1200.0, | |
| "grad_norm": 0.005759325809776783, | |
| "learning_rate": 4.962399180850277e-05, | |
| "loss": 0.0007, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1300.0, | |
| "grad_norm": 0.00447023706510663, | |
| "learning_rate": 4.955273299787451e-05, | |
| "loss": 0.0007, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1400.0, | |
| "grad_norm": 0.004739093128591776, | |
| "learning_rate": 4.947535513144287e-05, | |
| "loss": 0.0007, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1500.0, | |
| "grad_norm": 0.002100145909935236, | |
| "learning_rate": 4.939187749339435e-05, | |
| "loss": 0.0007, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1600.0, | |
| "grad_norm": 0.0027320764493197203, | |
| "learning_rate": 4.930232088810642e-05, | |
| "loss": 0.0007, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1700.0, | |
| "grad_norm": 0.0018412125064060092, | |
| "learning_rate": 4.920670763496268e-05, | |
| "loss": 0.0007, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1800.0, | |
| "grad_norm": 0.005061604082584381, | |
| "learning_rate": 4.910506156279029e-05, | |
| "loss": 0.0007, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1900.0, | |
| "grad_norm": 0.002439927076920867, | |
| "learning_rate": 4.899740800392143e-05, | |
| "loss": 0.0006, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2000.0, | |
| "grad_norm": 0.0013350448571145535, | |
| "learning_rate": 4.8883773787879873e-05, | |
| "loss": 0.0006, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2100.0, | |
| "grad_norm": 0.000241457688389346, | |
| "learning_rate": 4.8764187234694575e-05, | |
| "loss": 0.0006, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2200.0, | |
| "grad_norm": 0.0016053072176873684, | |
| "learning_rate": 4.863867814784168e-05, | |
| "loss": 0.0006, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2300.0, | |
| "grad_norm": 0.0009735923958942294, | |
| "learning_rate": 4.850727780681689e-05, | |
| "loss": 0.0006, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2400.0, | |
| "grad_norm": 0.002194364555180073, | |
| "learning_rate": 4.837001895933996e-05, | |
| "loss": 0.0006, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2500.0, | |
| "grad_norm": 0.0031507734674960375, | |
| "learning_rate": 4.822693581319326e-05, | |
| "loss": 0.0006, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2600.0, | |
| "grad_norm": 0.002784797688946128, | |
| "learning_rate": 4.8078064027696525e-05, | |
| "loss": 0.0006, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2700.0, | |
| "grad_norm": 0.0012538842856884003, | |
| "learning_rate": 4.792344070481972e-05, | |
| "loss": 0.0006, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2800.0, | |
| "grad_norm": 0.003489639377221465, | |
| "learning_rate": 4.7763104379936555e-05, | |
| "loss": 0.0006, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2900.0, | |
| "grad_norm": 0.0031125727109611034, | |
| "learning_rate": 4.75970950122206e-05, | |
| "loss": 0.0006, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 3000.0, | |
| "grad_norm": 0.0032254031393676996, | |
| "learning_rate": 4.742545397468659e-05, | |
| "loss": 0.0006, | |
| "step": 3000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 20000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20000, | |
| "save_steps": 500, | |
| "total_flos": 3.3457572937728e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |