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

    Neuronpedia

    INDEX
    Explanations

    configuration settings and defaults

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     azules
    0.43
     Zustand
    0.41
     조건
    0.41
     karan
    0.40
    맑
    0.40
     క్రికె
    0.40
     condi
    0.39
     condition
    0.39
     Teil
    0.39
     meydana
    0.39
    POSITIVE LOGITS
    default
    0.73
     default
    0.65
    Default
    0.64
     Default
    0.64
    デフォルト
    0.64
    CFBundle
    0.63
    max
    0.63
    Maximum
    0.63
     maximum
    0.63
    默认
    0.63
    Act Density 0.027%

    No Known Activations