© 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. 85282
    Prev
    Next
    INDEX
    Explanations

    publications and media types

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

    The neuron is strongly activated by words that name or label segments, media formats, or document sections (e.g. “show,” “podcast,” “segment,” “chat,” “blog,” “dissertation,” “proposal,” “wiki,” 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_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
     of
    -1.94
     in
    -1.66
    </h4>
    -1.30
     is
    -1.29
     to
    -1.28
    in
    -1.22
    。
    -1.20
    of
    -1.11
     about
    -1.10
    for
    -1.09
    POSITIVE LOGITS
     then
    1.45
    1.45
    urlencoded
    1.44
    +"'
    1.38
    Ѕ
    1.36
    1.36
    issipp
    1.32
    1.32
    tesse
    1.31
    莛
    1.30
    Activations Density 0.243%

    No Known Activations