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

    Neuronpedia

    INDEX
    Explanations

    technical features, particularly ones associated with security and software functions

    oai_token-act-pair · gpt-3.5-turbo

    technical terms related to flaws, features, and abilities in software or game mechanics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Books
    -0.78
    inas
    -0.76
    TeX
    -0.74
    adow
    -0.73
    endars
    -0.69
     Matters
    -0.67
    Topics
    -0.66
    Frames
    -0.66
    isitions
    -0.65
     Classics
    -0.65
    POSITIVE LOGITS
     whereby
    0.92
     nicknamed
    0.79
     applied
    0.78
     called
    0.74
     applies
    0.72
     workaround
    0.72
     known
    0.72
     allows
    0.67
     loophole
    0.67
     dubbed
    0.66
    Act Density 0.270%

    No Known Activations