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

    Neuronpedia

    INDEX
    Explanations

    corruption/broken

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     corrupted
    -0.71
    UIControlState
    -0.70
     détru
    -0.66
    fromnode
    -0.65
    zzleHttp
    -0.65
     distorted
    -0.65
     perverted
    -0.65
     dramatique
    -0.65
     frères
    -0.64
     decorada
    -0.63
    POSITIVE LOGITS
    s
    0.63
    sheet
    0.62
    san
    0.61
    CompilerServices
    0.59
    ity
    0.56
    sch
    0.56
    ness
    0.55
    ities
    0.50
    tling
    0.50
    space
    0.49
    Act Density 1.587%

    No Known Activations