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

    Neuronpedia

    INDEX
    Explanations

    harmed networks

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     harm
    -1.23
     injury
    -0.82
    nets
    -0.81
     harming
    -0.81
     Harm
    -0.80
     harmed
    -0.80
     Nets
    -0.79
     nets
    -0.76
     harms
    -0.74
    Nets
    -0.69
    POSITIVE LOGITS
    XtraReports
    0.81
    PerformLayout
    0.80
     समीक्षाएं
    0.80
    WriteBarrier
    0.71
    TagMode
    0.69
    postsleuth
    0.65
    UnsafeEnabled
    0.63
    sizeCache
    0.63
    makeText
    0.62
    aarrggbb
    0.61
    Act Density 0.357%

    No Known Activations