© 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. 7450
    Prev
    Next
    INDEX
    Explanations

    audio/video media

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

    The neuron fires strongly on words and abbreviations referring to audio or visual media (e.g. “audio,” “visual,” “AV,” “audiovisual”).

    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
     beziehungs
    -1.84
    酹
    -1.70
    したり
    -1.63
     или
    -1.62
     一九
    -1.62
    replyreport
    -1.60
     where
    -1.58
     또는
    -1.57
    Otras
    -1.55
    Lokasi
    -1.52
    POSITIVE LOGITS
    )\\
    1.82
    </strong>
    1.78
    鹇
    1.73
    \\
    1.68
    :\\
    1.66
    unii
    1.66
     sont
    1.63
    菒
    1.59
     purpoſe
    1.56
     opr
    1.52
    Activations Density 0.017%

    No Known Activations