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

    Neuronpedia

    INDEX
    Explanations

    Single letters

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    =no
    -0.07
    캉
    -0.07
    进去
    -0.07
    ONSE
    -0.07
    河流
    -0.07
    mund
    -0.06
    wall
    -0.06
    戕
    -0.06
    夺
    -0.06
    蜓
    -0.06
    POSITIVE LOGITS
    Dt
    0.08
     Ptr
    0.07
     Variant
    0.07
    #get
    0.07
    以下是
    0.07
     sped
    0.07
    domains
    0.07
    /li
    0.07
     znal
    0.07
    情報を
    0.06
    Act Density 0.009%

    No Known Activations