© 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. 22-GEMMASCOPE-RES-131K
    7. 26482
    Prev
    Next
    INDEX
    Explanations

    question starters

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

    The neuron flags question or interrogative constructions—especially question-opening words (Who, Are, How, etc.) and their immediate function words.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    リスク
    -0.93
     ͡°)
    -0.90
    ѐ
    -0.90
    лость
    -0.88
    :
    -0.88
    癜
    -0.88
    sockets
    -0.87
    analyse
    -0.84
    endcsname
    -0.83
    盃
    -0.83
    POSITIVE LOGITS
     your
    1.22
     you
    1.18
    nande
    1.01
     any
    0.98
     comfortable
    0.96
     think
    0.93
     jesteś
    0.92
    准备
    0.91
    őben
    0.91
    Literatuur
    0.89
    Activations Density 0.010%

    No Known Activations