© 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. 15235
    Prev
    Next
    INDEX
    Explanations

    October, See, Dennis, Kenya

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

    The neuron responds to segments of hyphenated compound words (e.g. the “low-”, “non-”, “fat-” parts in “low-fat,” “non-fat,” “fat-free,” etc.).

    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
     darm
    -1.03
     comunión
    -0.86
     eventually
    -0.82
     مراسم
    -0.81
     تحميل
    -0.81
     opinie
    -0.79
     inscripción
    -0.79
     vang
    -0.79
     miller
    -0.77
    hwar
    -0.77
    POSITIVE LOGITS
     another
    1.23
    Another
    1.16
     тоже
    1.01
     Another
    0.97
    getMinutes
    0.95
    skr
    0.95
     natomiast
    0.94
    Приятного
    0.93
    Rgds
    0.92
    另一种
    0.91
    Activations Density 0.034%

    No Known Activations