© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 8737
    Prev
    Next
    INDEX
    Explanations

    asking questions

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron fires on verbs that introduce questions or requests—especially “ask” (and its variants) and “questioned.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    走吧
    -1.59
     zowel
    -1.55
    亮相
    -1.44
    such
    -1.40
    ).
    -1.39
     takich
    -1.38
     reichlich
    -1.37
     mezclar
    -1.36
     geprü
    -1.34
     valorar
    -1.34
    POSITIVE LOGITS
     which
    2.52
     where
    1.91
     in
    1.68
    which
    1.65
     if
    1.54
     apakah
    1.42
    ции
    1.34
    capulco
    1.34
     russo
    1.33
     yPos
    1.33
    Activations Density 0.052%

    No Known Activations