unlearning-cleanslate/forget-pool-qwen3-8b

该数据集包含93个训练样本，每个样本具有多个特征，包括内容ID（content_id）、内容标题（content_title）、记忆分数（memorized_fraction）、最大概率值（max_p_z）、窗口数量（num_windows）、记忆窗口数量（memorized_windows）、来源数据集（source_dataset）、池化分箱（pool_bin）以及参考目标（reference_target）。数据集未明确说明具体用途，但特征表明可能与内容记忆或分析相关。

This dataset contains 93 training examples, each with multiple features including content ID (content_id), content title (content_title), memorized fraction (memorized_fraction), maximum probability value (max_p_z), number of windows (num_windows), memorized windows (memorized_windows), source dataset (source_dataset), pool bin (pool_bin), and reference target (reference_target). The dataset does not explicitly state its purpose, but the features suggest it may be related to content memorization or analysis.