| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9999310225901017, | |
| "eval_steps": 500, | |
| "global_step": 14497, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1.9771431199411332e-05, | |
| "loss": 3.0982, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.9541482707873437e-05, | |
| "loss": 1.5591, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 1.9311534216335545e-05, | |
| "loss": 1.1593, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 1.9081585724797646e-05, | |
| "loss": 0.979, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 1.8851637233259754e-05, | |
| "loss": 0.9437, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 1.8621688741721855e-05, | |
| "loss": 0.8984, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 1.8392200147167038e-05, | |
| "loss": 0.8578, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 1.816225165562914e-05, | |
| "loss": 0.8538, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1.7932303164091244e-05, | |
| "loss": 0.824, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 1.770235467255335e-05, | |
| "loss": 0.7844, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 1.7472406181015452e-05, | |
| "loss": 0.7969, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 1.7242917586460636e-05, | |
| "loss": 0.7902, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.7012969094922737e-05, | |
| "loss": 0.7709, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.6783020603384845e-05, | |
| "loss": 0.7573, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.1905351877212524, | |
| "eval_runtime": 298.8537, | |
| "eval_samples_per_second": 39.782, | |
| "eval_steps_per_second": 13.261, | |
| "step": 7248 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 1.655307211184695e-05, | |
| "loss": 0.603, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 1.632312362030905e-05, | |
| "loss": 0.4874, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 1.6093175128771158e-05, | |
| "loss": 0.4762, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 1.586322663723326e-05, | |
| "loss": 0.496, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 1.5633738042678443e-05, | |
| "loss": 0.4922, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 1.5403789551140547e-05, | |
| "loss": 0.4737, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 1.517384105960265e-05, | |
| "loss": 0.481, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 1.4943892568064754e-05, | |
| "loss": 0.485, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 1.4713944076526859e-05, | |
| "loss": 0.4892, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 1.448445548197204e-05, | |
| "loss": 0.4775, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 1.4254966887417219e-05, | |
| "loss": 0.4676, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 1.4025018395879325e-05, | |
| "loss": 0.4688, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 1.379506990434143e-05, | |
| "loss": 0.4785, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 1.3565121412803532e-05, | |
| "loss": 0.4734, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.368137240409851, | |
| "eval_runtime": 299.2167, | |
| "eval_samples_per_second": 39.734, | |
| "eval_steps_per_second": 13.245, | |
| "step": 14497 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 43488, | |
| "num_train_epochs": 6, | |
| "save_steps": 500, | |
| "total_flos": 1.135274382017962e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |