© 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. 34-GEMMASCOPE-RES-131K
    7. 47543
    Prev
    Next
    INDEX
    Explanations

    my, photo, Color, The, In followed by common pairings

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

    The neuron flags actual user‐written post content (the substantive message text) rather than the preceding boilerplate or navigation boilerplate.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    lula
    -1.20
    kva
    -1.09
    穿
    -0.95
     fave
    -0.94
    alka
    -0.94
     yamaha
    -0.94
     toshiba
    -0.93
    盼
    -0.93
    tecla
    -0.92
     chevrolet
    -0.92
    POSITIVE LOGITS
     =\
    1.20
     OEM
    1.08
     moletom
    1.06
    findpost
    1.06
     Motorsport
    1.06
    Gön
    1.02
     maxima
    1.01
    TÜ
    1.01
     frappé
    1.01
     Motorsports
    1.00
    Activations Density 0.059%

    No Known Activations