© 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. 31049
    Prev
    Next
    INDEX
    Explanations

    Pathfinder game rules

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

    This neuron is specialized for spotting numeric tokens or numbers (including digits, numeric references, and numbers in parentheses).

    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
     空間
    -0.75
    Gé
    -0.73
     Sailor
    -0.71
    喻
    -0.70
    sdp
    -0.69
     См
    -0.69
     Giver
    -0.69
    邬
    -0.69
    意味
    -0.68
    ใหญ
    -0.68
    POSITIVE LOGITS
     Oregon
    0.93
    Oregon
    0.88
     OREGON
    0.82
     Boise
    0.82
     Torino
    0.79
    小时
    0.75
    레이
    0.73
     attendees
    0.73
     Escuela
    0.71
     Paz
    0.69
    Activations Density 0.015%

    No Known Activations