{ "best_metric": 35.4918, "best_model_checkpoint": "bin/liputan6-base/checkpoint-36354", "epoch": 3.0, "eval_steps": 500, "global_step": 36354, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "grad_norm": 0.8706055879592896, "learning_rate": 0.0006666666666666666, "loss": 2.0683, "step": 12118 }, { "epoch": 1.0, "eval_gen_len": 38.11319722931098, "eval_loss": 2.372281312942505, "eval_rouge1": 33.423, "eval_rouge2": 16.5875, "eval_rougeL": 28.0723, "eval_rougeLsum": 31.0371, "eval_runtime": 1335.7227, "eval_samples_per_second": 8.214, "eval_steps_per_second": 0.129, "step": 12118 }, { "epoch": 2.0, "grad_norm": 0.7888175845146179, "learning_rate": 0.0003333333333333333, "loss": 1.5342, "step": 24236 }, { "epoch": 2.0, "eval_gen_len": 36.71764491432738, "eval_loss": 2.139983654022217, "eval_rouge1": 34.4775, "eval_rouge2": 17.1757, "eval_rougeL": 28.8546, "eval_rougeLsum": 32.0264, "eval_runtime": 1116.2631, "eval_samples_per_second": 9.829, "eval_steps_per_second": 0.154, "step": 24236 }, { "epoch": 3.0, "grad_norm": 0.6657369136810303, "learning_rate": 0.0, "loss": 1.241, "step": 36354 }, { "epoch": 3.0, "eval_gen_len": 36.4740247903755, "eval_loss": 2.039839029312134, "eval_rouge1": 35.4918, "eval_rouge2": 17.885, "eval_rougeL": 29.6804, "eval_rougeLsum": 32.9507, "eval_runtime": 1054.5787, "eval_samples_per_second": 10.404, "eval_steps_per_second": 0.163, "step": 36354 }, { "epoch": 3.0, "step": 36354, "total_flos": 3.9828842929402675e+17, "train_loss": 1.6144965134510645, "train_runtime": 36481.981, "train_samples_per_second": 15.943, "train_steps_per_second": 0.996 } ], "logging_steps": 500, "max_steps": 36354, "num_input_tokens_seen": 0, "num_train_epochs": 3, "save_steps": 500, "total_flos": 3.9828842929402675e+17, "train_batch_size": 16, "trial_name": null, "trial_params": null }