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

    Neuronpedia

    INDEX
    Explanations

    poorly understood mechanisms

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    _k
    -0.07
    .writeValue
    -0.06
     newcomer
    -0.06
    \"
    -0.06
    _As
    -0.06
    uxt
    -0.06
    upil
    -0.06
    fortunate
    -0.06
     expiresIn
    -0.06
    ξη
    -0.06
    POSITIVE LOGITS
     Photos
    0.07
    .Call
    0.06
    莱
    0.06
    坏
    0.06
     〜
    0.06
     destinationViewController
    0.06
    -focus
    0.06
     decom
    0.06
     oak
    0.06
    폴
    0.06
    Act Density 0.027%

    No Known Activations