{ "best_metric": 0.09211140125989914, "best_model_checkpoint": "/kaggle/working/checkpoints/checkpoint-550", "epoch": 3.0, "eval_steps": 50, "global_step": 558, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.26881720430107525, "eval_loss": 0.8808531165122986, "eval_runtime": 262.56, "eval_samples_per_second": 5.039, "eval_steps_per_second": 0.316, "step": 50 }, { "epoch": 0.5376344086021505, "grad_norm": 1.2331559658050537, "learning_rate": 0.0002, "loss": 1.3353, "step": 100 }, { "epoch": 0.5376344086021505, "eval_loss": 0.34334027767181396, "eval_runtime": 260.9751, "eval_samples_per_second": 5.069, "eval_steps_per_second": 0.318, "step": 100 }, { "epoch": 0.8064516129032258, "eval_loss": 0.16361241042613983, "eval_runtime": 260.6893, "eval_samples_per_second": 5.075, "eval_steps_per_second": 0.318, "step": 150 }, { "epoch": 1.075268817204301, "grad_norm": 1.0327280759811401, "learning_rate": 0.00017738257275395404, "loss": 0.1586, "step": 200 }, { "epoch": 1.075268817204301, "eval_loss": 0.1293419897556305, "eval_runtime": 260.7002, "eval_samples_per_second": 5.075, "eval_steps_per_second": 0.318, "step": 200 }, { "epoch": 1.3440860215053765, "eval_loss": 0.11591021716594696, "eval_runtime": 258.6801, "eval_samples_per_second": 5.114, "eval_steps_per_second": 0.321, "step": 250 }, { "epoch": 1.6129032258064515, "grad_norm": 0.838832676410675, "learning_rate": 0.00011976125132041974, "loss": 0.0929, "step": 300 }, { "epoch": 1.6129032258064515, "eval_loss": 0.10916169732809067, "eval_runtime": 260.2725, "eval_samples_per_second": 5.083, "eval_steps_per_second": 0.319, "step": 300 }, { "epoch": 1.881720430107527, "eval_loss": 0.10310520976781845, "eval_runtime": 230.3193, "eval_samples_per_second": 5.744, "eval_steps_per_second": 0.36, "step": 350 }, { "epoch": 2.150537634408602, "grad_norm": 0.7198598980903625, "learning_rate": 5.3200956606277006e-05, "loss": 0.0794, "step": 400 }, { "epoch": 2.150537634408602, "eval_loss": 0.09767071157693863, "eval_runtime": 226.6934, "eval_samples_per_second": 5.836, "eval_steps_per_second": 0.366, "step": 400 }, { "epoch": 2.4193548387096775, "eval_loss": 0.09671153873205185, "eval_runtime": 227.5066, "eval_samples_per_second": 5.815, "eval_steps_per_second": 0.365, "step": 450 }, { "epoch": 2.688172043010753, "grad_norm": 0.5379258990287781, "learning_rate": 7.810141074975818e-06, "loss": 0.0711, "step": 500 }, { "epoch": 2.688172043010753, "eval_loss": 0.093271903693676, "eval_runtime": 228.075, "eval_samples_per_second": 5.801, "eval_steps_per_second": 0.364, "step": 500 }, { "epoch": 2.956989247311828, "eval_loss": 0.09211140125989914, "eval_runtime": 229.0579, "eval_samples_per_second": 5.776, "eval_steps_per_second": 0.362, "step": 550 } ], "logging_steps": 100, "max_steps": 558, "num_input_tokens_seen": 0, "num_train_epochs": 3, "save_steps": 50, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 1.545337824120668e+17, "train_batch_size": 16, "trial_name": null, "trial_params": null }