| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.21999853334311104, | |
| "eval_steps": 5500, | |
| "global_step": 33000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 9.997777792592494e-05, | |
| "loss": 0.0307, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 9.995555585184987e-05, | |
| "loss": 0.0312, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 9.993333377777482e-05, | |
| "loss": 0.0311, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 9.991111170369976e-05, | |
| "loss": 0.0309, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 9.988888962962469e-05, | |
| "loss": 0.0311, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 9.986666755554963e-05, | |
| "loss": 0.0309, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 9.984444548147457e-05, | |
| "loss": 0.031, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 9.982222340739951e-05, | |
| "loss": 0.0309, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 9.980000133332444e-05, | |
| "loss": 0.0312, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 9.977777925924938e-05, | |
| "loss": 0.031, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 9.975555718517432e-05, | |
| "loss": 0.031, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_loss": 0.031031738966703415, | |
| "eval_runtime": 12112.511, | |
| "eval_samples_per_second": 2.064, | |
| "eval_steps_per_second": 0.344, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 9.973333511109927e-05, | |
| "loss": 0.0308, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 9.971111303702421e-05, | |
| "loss": 0.0311, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.968889096294914e-05, | |
| "loss": 0.031, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.966666888887408e-05, | |
| "loss": 0.0313, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.964444681479902e-05, | |
| "loss": 0.0311, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 9.962222474072396e-05, | |
| "loss": 0.031, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 9.960000266664889e-05, | |
| "loss": 0.0313, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 9.957778059257383e-05, | |
| "loss": 0.0313, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.955555851849878e-05, | |
| "loss": 0.031, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.953333644442371e-05, | |
| "loss": 0.0318, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.951111437034864e-05, | |
| "loss": 0.0316, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 0.030834877863526344, | |
| "eval_runtime": 12107.9455, | |
| "eval_samples_per_second": 2.065, | |
| "eval_steps_per_second": 0.344, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 9.948889229627358e-05, | |
| "loss": 0.0311, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 9.946667022219853e-05, | |
| "loss": 0.0308, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 9.944444814812346e-05, | |
| "loss": 0.0308, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 9.94222260740484e-05, | |
| "loss": 0.0309, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 9.940000399997333e-05, | |
| "loss": 0.0308, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 9.937778192589828e-05, | |
| "loss": 0.0313, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.935555985182321e-05, | |
| "loss": 0.0311, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.933333777774815e-05, | |
| "loss": 0.0312, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.93111157036731e-05, | |
| "loss": 0.0311, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 9.928889362959803e-05, | |
| "loss": 0.031, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 9.926667155552298e-05, | |
| "loss": 0.031, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "eval_loss": 0.03085586242377758, | |
| "eval_runtime": 12089.5064, | |
| "eval_samples_per_second": 2.068, | |
| "eval_steps_per_second": 0.345, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 9.924444948144791e-05, | |
| "loss": 0.0309, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.922222740737285e-05, | |
| "loss": 0.0311, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.920000533329778e-05, | |
| "loss": 0.0315, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.917778325922273e-05, | |
| "loss": 0.0308, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 9.915556118514766e-05, | |
| "loss": 0.0315, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 9.91333391110726e-05, | |
| "loss": 0.0309, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 9.911111703699753e-05, | |
| "loss": 0.0313, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.908889496292248e-05, | |
| "loss": 0.031, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.906667288884741e-05, | |
| "loss": 0.0311, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.904445081477235e-05, | |
| "loss": 0.0314, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 9.902222874069728e-05, | |
| "loss": 0.0311, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 0.030636409297585487, | |
| "eval_runtime": 12087.8542, | |
| "eval_samples_per_second": 2.068, | |
| "eval_steps_per_second": 0.345, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 9.900000666662223e-05, | |
| "loss": 0.0311, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 9.897778459254717e-05, | |
| "loss": 0.0311, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.89555625184721e-05, | |
| "loss": 0.0307, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.893334044439703e-05, | |
| "loss": 0.0308, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.891111837032197e-05, | |
| "loss": 0.0306, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 9.888889629624692e-05, | |
| "loss": 0.031, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 9.886667422217187e-05, | |
| "loss": 0.0311, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 9.88444521480968e-05, | |
| "loss": 0.0307, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 9.882223007402173e-05, | |
| "loss": 0.0309, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 9.880000799994667e-05, | |
| "loss": 0.0308, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 9.877778592587162e-05, | |
| "loss": 0.0308, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_loss": 0.030515629798173904, | |
| "eval_runtime": 12091.2293, | |
| "eval_samples_per_second": 2.068, | |
| "eval_steps_per_second": 0.345, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 9.875556385179655e-05, | |
| "loss": 0.0313, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 9.873334177772149e-05, | |
| "loss": 0.031, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 9.871111970364642e-05, | |
| "loss": 0.0311, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 9.868889762957137e-05, | |
| "loss": 0.031, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 9.86666755554963e-05, | |
| "loss": 0.0312, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 9.864445348142124e-05, | |
| "loss": 0.0307, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 9.862223140734617e-05, | |
| "loss": 0.0309, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 9.860000933327112e-05, | |
| "loss": 0.0308, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 9.857778725919605e-05, | |
| "loss": 0.0309, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 9.855556518512099e-05, | |
| "loss": 0.0311, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 9.853334311104592e-05, | |
| "loss": 0.0307, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 0.030391648411750793, | |
| "eval_runtime": 12101.6748, | |
| "eval_samples_per_second": 2.066, | |
| "eval_steps_per_second": 0.344, | |
| "step": 33000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2250015, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "save_steps": 5500, | |
| "total_flos": 1.714489493815296e+18, | |
| "train_batch_size": 6, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |