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

    Neuronpedia

    INDEX
    Explanations

    corrupt

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    ěž
    -0.07
     dojo
    -0.07
     sentinel
    -0.06
     contro
    -0.06
    -ios
    -0.06
    allery
    -0.06
    ("");↵
    -0.06
    indows
    -0.06
     �
    -0.06
     Silence
    -0.06
    POSITIVE LOGITS
     corruption
    0.08
     Corruption
    0.08
     corrupt
    0.08
     Salmon
    0.07
    ROT
    0.07
    rupted
    0.06
    ков
    0.06
     Corner
    0.06
     bankrupt
    0.06
    egrity
    0.06
    Act Density 0.004%

    No Known Activations