Quiz collegato alla guida · Difficile Livello

Codificatori automatici sparsi per il quiz sull'estrazione delle funzionalità

Studia dizionari e ricostruzioni di caratteristiche sparse, con l'esperimento del Golden Gate Bridge, i latenti sovracompleti, lo sterzo e i confini del modello.

Domanda 1 di 8

Which interpretability difficulty motivates sparse dictionary learning for neural activations?

Anteprima di ogni domanda8 domande

Le risposte e le spiegazioni vengono visualizzate mentre rispondi al quiz.

  1. Which interpretability difficulty motivates sparse dictionary learning for neural activations?
    • 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. What does an SAE sparsity constraint encourage for a given input?
    • 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. Nel lavoro "Scaling Monosemanticity" di Anthropic, qual era la famosa caratteristica di esempio?
    • La caratteristica del Golden Gate Bridge
    • Una caratteristica per il numero sette
    • Una caratteristica solo per il colore blu
    • Una caratteristica per le vocali
  4. What does an overcomplete SAE dictionary mean?
    • 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. What does a monosemantic feature label aim to describe in this research context?
    • 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. Come possono essere utilizzati i SAE per influenzare il comportamento di un modello?
    • Attivando o disattivando una funzione per gestire le uscite
    • Riqualificando l'intero modello da zero
    • Eliminando il tokenizer
    • Convertendolo in un modello di immagine
  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