{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.06033182503770739, "eval_steps": 10, "global_step": 50, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0012066365007541479, "eval_loss": 2.083979368209839, "eval_runtime": 24.3217, "eval_samples_per_second": 14.349, "eval_steps_per_second": 7.195, "step": 1 }, { "epoch": 0.006033182503770739, "grad_norm": 1.081383228302002, "learning_rate": 5e-05, "loss": 2.0649, "step": 5 }, { "epoch": 0.012066365007541479, "grad_norm": 1.181739330291748, "learning_rate": 0.0001, "loss": 1.892, "step": 10 }, { "epoch": 0.012066365007541479, "eval_loss": 1.8972073793411255, "eval_runtime": 24.5365, "eval_samples_per_second": 14.224, "eval_steps_per_second": 7.132, "step": 10 }, { "epoch": 0.01809954751131222, "grad_norm": 1.176864504814148, "learning_rate": 9.619397662556435e-05, "loss": 1.8032, "step": 15 }, { "epoch": 0.024132730015082957, "grad_norm": 0.9331520199775696, "learning_rate": 8.535533905932738e-05, "loss": 1.7319, "step": 20 }, { "epoch": 0.024132730015082957, "eval_loss": 1.652181625366211, "eval_runtime": 24.6211, "eval_samples_per_second": 14.175, "eval_steps_per_second": 7.108, "step": 20 }, { "epoch": 0.030165912518853696, "grad_norm": 0.9450703263282776, "learning_rate": 6.91341716182545e-05, "loss": 1.4481, "step": 25 }, { "epoch": 0.03619909502262444, "grad_norm": 0.8512829542160034, "learning_rate": 5e-05, "loss": 1.5354, "step": 30 }, { "epoch": 0.03619909502262444, "eval_loss": 1.5946625471115112, "eval_runtime": 24.7054, "eval_samples_per_second": 14.126, "eval_steps_per_second": 7.083, "step": 30 }, { "epoch": 0.042232277526395176, "grad_norm": 1.084594964981079, "learning_rate": 3.086582838174551e-05, "loss": 1.6278, "step": 35 }, { "epoch": 0.048265460030165915, "grad_norm": 0.8652676939964294, "learning_rate": 1.4644660940672627e-05, "loss": 1.539, "step": 40 }, { "epoch": 0.048265460030165915, "eval_loss": 1.575426459312439, "eval_runtime": 25.0271, "eval_samples_per_second": 13.945, "eval_steps_per_second": 6.992, "step": 40 }, { "epoch": 0.05429864253393665, "grad_norm": 0.9328100681304932, "learning_rate": 3.8060233744356633e-06, "loss": 1.5162, "step": 45 }, { "epoch": 0.06033182503770739, "grad_norm": 1.0344215631484985, "learning_rate": 0.0, "loss": 1.5315, "step": 50 }, { "epoch": 0.06033182503770739, "eval_loss": 1.5716311931610107, "eval_runtime": 25.008, "eval_samples_per_second": 13.956, "eval_steps_per_second": 6.998, "step": 50 } ], "logging_steps": 5, "max_steps": 50, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 13, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 9432777665544192.0, "train_batch_size": 2, "trial_name": null, "trial_params": null }