© 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. 72546
    Prev
    Next
    INDEX
    Explanations

    WIN, SIR, BUY, Côte

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

    The neuron fires on tokens beginning with “WIN” (as in Windows version or platform markers).

    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
    は
    -3.00
    In
    -2.97
    d
    -2.78
    で
    -2.69
    er
    -2.66
    These
    -2.52
    If
    -2.50
    unteering
    -2.44
    o
    -2.42
    K
    -2.39
    POSITIVE LOGITS
    埽
    3.55
    栳
    3.00
    合适的
    2.98
    ing
    2.92
    筇
    2.80
    塒
    2.78
    鯪
    2.56
    质感
    2.53
    ίες
    2.48
    㩒
    2.47
    Activations Density 0.003%

    No Known Activations