| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.796623177283193, | |
| "eval_steps": 500, | |
| "global_step": 25000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.9616270145817347e-05, | |
| "loss": 0.2232, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 1.9232540291634693e-05, | |
| "loss": 0.1483, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 1.8848810437452035e-05, | |
| "loss": 0.1446, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 1.846508058326938e-05, | |
| "loss": 0.1343, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 1.8081350729086726e-05, | |
| "loss": 0.1195, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 1.7697620874904068e-05, | |
| "loss": 0.1209, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 1.7313891020721413e-05, | |
| "loss": 0.1186, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 1.693016116653876e-05, | |
| "loss": 0.0884, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 1.65464313123561e-05, | |
| "loss": 0.1025, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 1.6162701458173446e-05, | |
| "loss": 0.1051, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.11643879860639572, | |
| "eval_runtime": 429.8512, | |
| "eval_samples_per_second": 12.125, | |
| "eval_steps_per_second": 3.031, | |
| "step": 5212 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 1.5778971603990792e-05, | |
| "loss": 0.0896, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 1.5395241749808137e-05, | |
| "loss": 0.0864, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 1.5011511895625481e-05, | |
| "loss": 0.0913, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 1.4627782041442827e-05, | |
| "loss": 0.0857, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 1.4244052187260169e-05, | |
| "loss": 0.0853, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 1.3860322333077514e-05, | |
| "loss": 0.082, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 1.3476592478894858e-05, | |
| "loss": 0.0887, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 1.3092862624712203e-05, | |
| "loss": 0.0797, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 1.2709132770529549e-05, | |
| "loss": 0.0805, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 1.2325402916346893e-05, | |
| "loss": 0.0717, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.08999563753604889, | |
| "eval_runtime": 429.7074, | |
| "eval_samples_per_second": 12.129, | |
| "eval_steps_per_second": 3.032, | |
| "step": 10424 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 1.1941673062164237e-05, | |
| "loss": 0.0714, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 1.155794320798158e-05, | |
| "loss": 0.0674, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 1.1174213353798926e-05, | |
| "loss": 0.0626, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 1.0790483499616271e-05, | |
| "loss": 0.0705, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 1.0406753645433617e-05, | |
| "loss": 0.0702, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 1.002302379125096e-05, | |
| "loss": 0.064, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 9.639293937068304e-06, | |
| "loss": 0.066, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 9.25556408288565e-06, | |
| "loss": 0.0726, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 8.871834228702994e-06, | |
| "loss": 0.0668, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 8.48810437452034e-06, | |
| "loss": 0.0728, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 8.104374520337683e-06, | |
| "loss": 0.0619, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.08260077238082886, | |
| "eval_runtime": 429.8218, | |
| "eval_samples_per_second": 12.126, | |
| "eval_steps_per_second": 3.031, | |
| "step": 15636 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 7.720644666155027e-06, | |
| "loss": 0.0697, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 7.336914811972372e-06, | |
| "loss": 0.0552, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 6.953184957789717e-06, | |
| "loss": 0.0568, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 6.569455103607061e-06, | |
| "loss": 0.0596, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 6.185725249424405e-06, | |
| "loss": 0.0551, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 5.801995395241751e-06, | |
| "loss": 0.0608, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 5.418265541059095e-06, | |
| "loss": 0.0572, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 5.034535686876439e-06, | |
| "loss": 0.0488, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 4.650805832693784e-06, | |
| "loss": 0.0626, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 4.2670759785111286e-06, | |
| "loss": 0.0528, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.08744284510612488, | |
| "eval_runtime": 429.0419, | |
| "eval_samples_per_second": 12.148, | |
| "eval_steps_per_second": 3.037, | |
| "step": 20848 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 3.883346124328473e-06, | |
| "loss": 0.059, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 3.4996162701458174e-06, | |
| "loss": 0.0516, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 3.1158864159631625e-06, | |
| "loss": 0.0455, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "learning_rate": 2.7321565617805067e-06, | |
| "loss": 0.0515, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 2.3484267075978514e-06, | |
| "loss": 0.0505, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "learning_rate": 1.964696853415196e-06, | |
| "loss": 0.0468, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "learning_rate": 1.5809669992325404e-06, | |
| "loss": 0.0464, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 1.197237145049885e-06, | |
| "loss": 0.0429, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 8.135072908672295e-07, | |
| "loss": 0.0527, | |
| "step": 25000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 26060, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 4.50113978400768e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |