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

    Neuronpedia

    INDEX
    Explanations

    phrases related to boundaries being crossed

    oai_token-act-pair · gpt-3.5-turbo

    references to crossing boundaries or lines, both literal and metaphorical

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    thood
    -0.95
    LESS
    -0.74
    FILE
    -0.71
    FORE
    -0.69
    DAY
    -0.67
    summary
    -0.66
    ummies
    -0.64
     alike
    -0.64
    manship
    -0.63
    larg
    -0.63
    POSITIVE LOGITS
     threshold
    1.35
     boundary
    1.03
     boundaries
    1.02
     thresholds
    1.02
     wire
    0.91
     border
    0.90
     hurdle
    0.88
     bounds
    0.87
     fence
    0.87
     limit
    0.86
    Act Density 0.102%

    No Known Activations