© 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. 22469
    Prev
    Next
    INDEX
    Explanations

    sports and fencing types

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

    The neuron fires strongly on mentions of sports—especially major competition names like Wimbledon and the sport term fencing.

    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
    亖
    -2.47
    Meik
    -2.05
    '
    -2.02
     видим
    -1.99
     bankası
    -1.97
    लेट
    -1.84
     önem
    -1.80
     embellished
    -1.76
     correta
    -1.75
    自然的
    -1.73
    POSITIVE LOGITS
    ."
    2.34
    ка
    2.30
     You
    2.28
    ни
    2.08
    the
    2.02
    i
    2.00
    ");
    1.99
     самый
    1.99
    ран
    1.97
    m
    1.95
    Activations Density 0.002%

    No Known Activations