引導式測驗 · 硬 等級
用於特徵提取測驗的稀疏自動編碼器
透過金門大橋實驗、超完備潛伏、轉向和模型邊界,研究稀疏特徵字典和重建。
問題 1 的 8
哪種可解釋性困難促成了神經激活的稀疏字典學習?
預覽每個問題8 問題
當您參加測驗時,就會出現答案和解釋。
- 哪種可解釋性困難促成了神經激活的稀疏字典學習?
- 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
- SAE 稀疏性限制對給定輸入有何促進作用?
- 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
- 在 Anthropic 的「Scaling Monosemanticity」工作中,著名的範例功能是什麼?
- 金門大橋特色
- 數字七的特徵
- 僅適用於藍色的功能
- 元音的特徵
- 超完備(overcomplete)的 SAE 字典是什麼意思?
- 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
- 在這個研究情境中,單義特徵標籤旨在描述什麼?
- 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
- 如何使用 SAE 來影響模型的行為?
- 透過開啟或關閉某個功能來控制輸出
- 透過從頭開始重新訓練整個模型
- 透過刪除標記器
- 透過將其轉換為圖像模型
- 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
- 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