Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    Reference IDs

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     first
    -0.07
     说
    -0.06
     DPI
    -0.06
    ensis
    -0.06
    pw
    -0.06
    žitě
    -0.06
     Nordic
    -0.06
    ですが
    -0.06
     socio
    -0.06
     March
    -0.06
    POSITIVE LOGITS
     CreateMap
    0.06
    []):
    0.06
    ornecedor
    0.06
    �
    0.06
    relude
    0.06
     통
    0.06
     eql
    0.06
     ATA
    0.06
    TokenType
    0.06
    ossal
    0.06
    Act Density 0.013%

    No Known Activations