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

    Neuronpedia

    INDEX
    Explanations

    conditional sentences describing hypothetical scenarios

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating hypothetical scenarios or conditions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Dur
    -0.70
    Fine
    -0.66
     horizont
    -0.65
     DOS
    -0.64
     JPEG
    -0.62
     corrid
    -0.62
     Fiesta
    -0.62
     SVG
    -0.61
    Solid
    -0.61
     seams
    -0.61
    POSITIVE LOGITS
    wered
    1.13
    lled
    0.94
     recover
    0.86
     overcome
    0.85
     partake
    0.82
    plet
    0.82
    iling
    0.81
     survive
    0.80
    ogle
    0.80
    ying
    0.79
    Act Density 0.037%

    No Known Activations