{ "best_metric": null, "best_model_checkpoint": null, "epoch": 38.72884283246977, "eval_steps": 500, "global_step": 7000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 5.53, "learning_rate": 1.9800000000000004e-05, "loss": 8.0967, "step": 1000 }, { "epoch": 11.09, "learning_rate": 3.619444444444445e-05, "loss": 2.6161, "step": 2000 }, { "epoch": 11.09, "eval_bleu": 13.5109, "eval_gen_len": 19.1966, "eval_loss": 3.176215171813965, "eval_runtime": 298.8186, "eval_samples_per_second": 9.685, "eval_steps_per_second": 1.211, "step": 2000 }, { "epoch": 13.86, "eval_bleu": 16.2868, "eval_gen_len": 18.7985, "eval_loss": 3.037487268447876, "eval_runtime": 263.6174, "eval_samples_per_second": 10.978, "eval_steps_per_second": 1.373, "step": 2500 }, { "epoch": 16.62, "learning_rate": 2.9270833333333338e-05, "loss": 1.4467, "step": 3000 }, { "epoch": 16.62, "eval_bleu": 17.6991, "eval_gen_len": 18.1949, "eval_loss": 3.132826328277588, "eval_runtime": 253.8432, "eval_samples_per_second": 11.401, "eval_steps_per_second": 1.426, "step": 3000 }, { "epoch": 19.39, "eval_bleu": 17.9052, "eval_gen_len": 18.3117, "eval_loss": 3.2690258026123047, "eval_runtime": 250.4816, "eval_samples_per_second": 11.554, "eval_steps_per_second": 1.445, "step": 3500 }, { "epoch": 22.15, "learning_rate": 2.232638888888889e-05, "loss": 0.6809, "step": 4000 }, { "epoch": 22.15, "eval_bleu": 18.4075, "eval_gen_len": 18.2149, "eval_loss": 3.3850443363189697, "eval_runtime": 248.7531, "eval_samples_per_second": 11.634, "eval_steps_per_second": 1.455, "step": 4000 }, { "epoch": 24.91, "eval_bleu": 19.0339, "eval_gen_len": 18.009, "eval_loss": 3.446467876434326, "eval_runtime": 246.4301, "eval_samples_per_second": 11.744, "eval_steps_per_second": 1.469, "step": 4500 }, { "epoch": 27.68, "learning_rate": 1.5381944444444445e-05, "loss": 0.3422, "step": 5000 }, { "epoch": 27.68, "eval_bleu": 18.7281, "eval_gen_len": 17.5902, "eval_loss": 3.5680091381073, "eval_runtime": 242.436, "eval_samples_per_second": 11.937, "eval_steps_per_second": 1.493, "step": 5000 }, { "epoch": 30.44, "eval_bleu": 19.1534, "eval_gen_len": 18.2177, "eval_loss": 3.634960651397705, "eval_runtime": 247.9202, "eval_samples_per_second": 11.673, "eval_steps_per_second": 1.46, "step": 5500 }, { "epoch": 33.2, "learning_rate": 8.444444444444446e-06, "loss": 0.1941, "step": 6000 }, { "epoch": 33.2, "eval_bleu": 19.2575, "eval_gen_len": 17.8784, "eval_loss": 3.7152717113494873, "eval_runtime": 241.6987, "eval_samples_per_second": 11.974, "eval_steps_per_second": 1.498, "step": 6000 }, { "epoch": 35.97, "eval_bleu": 19.2475, "eval_gen_len": 17.9831, "eval_loss": 3.738178253173828, "eval_runtime": 245.2172, "eval_samples_per_second": 11.802, "eval_steps_per_second": 1.476, "step": 6500 }, { "epoch": 38.73, "learning_rate": 1.5e-06, "loss": 0.1271, "step": 7000 }, { "epoch": 38.73, "eval_bleu": 19.3045, "eval_gen_len": 17.9889, "eval_loss": 3.757262706756592, "eval_runtime": 245.3062, "eval_samples_per_second": 11.798, "eval_steps_per_second": 1.476, "step": 7000 } ], "logging_steps": 1000, "max_steps": 7200, "num_train_epochs": 40, "save_steps": 500, "total_flos": 1.941748272070656e+18, "trial_name": null, "trial_params": null }