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

    Neuronpedia

    INDEX
    Explanations

    code characters such as "this.", numbers, and specific variable names

    oai_token-act-pair · gemini-2.0-flash

    this, 0

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     hung
    -0.50
     مج
    -0.48
     ged
    -0.47
     M
    -0.47
     most
    -0.47
     Hanging
    -0.45
     hanging
    -0.44
     Ra
    -0.44
     conf
    -0.43
     vo
    -0.43
    POSITIVE LOGITS
     wireType
    0.86
    IntoConstraints
    0.82
    Personendaten
    0.82
     pleaſure
    0.79
    Tikang
    0.78
     kaarangay
    0.78
     houſe
    0.78
    +#+#
    0.76
    ConstraintMaker
    0.75
     ſtate
    0.75
    Act Density 0.062%

    No Known Activations