用于特征提取测验的稀疏自动编码器

当您参加测验时，就会出现答案和解释。

1. 哪种可解释性困难激发了神经激活的稀疏字典学习？
   • Every model activation is already a complete dictionary definition
   • Several learned signals can overlap in an activation space
   • The tokenizer must be permanently removed
   • All neural weights are stored as image pixels
2. 对于给定的输入，SAE 稀疏性约束会鼓励什么？
   • Every feature to activate equally on every input
   • A mandatory removal of its decoder
   • A guarantee that every feature has exactly one perfect human label
   • Only a limited subset of learned feature activations to be nonzero
3. 在 Anthropic 的“Scaling Monosemanticity”工作中，著名的示例功能是什么？
   • 金门大桥特色
   • 数字七的特征
   • 仅适用于蓝色的功能
   • 元音的特征
4. 超完备的 SAE 词典是什么意思？
   • It must activate every feature for every token
   • It contains no reconstruction mapping
   • It has more learned feature directions than input activation dimensions
   • It always requires less storage than the input vector
5. 在本研究背景下，单语义特征标签旨在描述什么？
   • A guarantee that the feature never responds to related examples
   • A feature associated with one interpretable concept
   • A language model that has only one parameter
   • A token represented by exactly one raw byte
6. 如何使用 SAE 来影响模型的行为？
   • 通过打开或关闭某个功能来控制输出
   • 通过从头开始重新训练整个模型
   • 通过删除标记器
   • 通过将其转换为图像模型
7. What does an SAE’s reconstruction residual measure for an activation vector?
   • The difference between two unrelated document titles
   • The total number of words in the model vocabulary
   • Only the model’s hardware temperature
   • The difference between the original activation and its decoded reconstruction
8. When an SAE is fitted to collected activations of a frozen language model, which weights need to be trained for that fitting?
   • Every original language-model weight must be updated
   • Only the tokenizer’s character list
   • The SAE’s own encoder and decoder weights
   • No parameters of the SAE can change