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

    Neuronpedia

    INDEX
    Explanations

    percentages in text

    oai_token-act-pair · gpt-3.5-turbo

    percentages and their related expressions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    worldly
    -0.92
    zie
    -0.82
    åĤ
    -0.75
    edin
    -0.74
    isure
    -0.73
    andr
    -0.70
    æ©
    -0.67
    achus
    -0.65
    ipples
    -0.64
    anse
    -0.64
    POSITIVE LOGITS
     Invisible
    0.91
     certainty
    0.84
     satisfaction
    0.80
     accuracy
    0.77
     pure
    0.76
     humidity
    0.76
     purity
    0.71
     accurate
    0.71
    heartedly
    0.69
     cotton
    0.69
    Act Density 0.047%

    No Known Activations