© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 14564
    Prev
    Next
    INDEX
    Explanations

    Daniel Radcliffe to star

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

    This neuron primarily activates on question-introducing words (like What, How, Why, Does) and related tokens in interrogative sentences.

    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
     LIABLE
    -1.79
     alarme
    -1.44
    "];
    -1.43
     nage
    -1.43
     coute
    -1.41
     malha
    -1.41
     député
    -1.40
    springen
    -1.38
    分享
    -1.38
     !!!!!
    -1.38
    POSITIVE LOGITS
     as
    2.09
     to
    2.08
     for
    1.91
     during
    1.88
     with
    1.85
     using
    1.79
     going
    1.76
     after
    1.70
     trying
    1.69
     begin
    1.68
    Activations Density 0.014%

    No Known Activations