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

    Neuronpedia

    INDEX
    Explanations

    phrases or sentences with a contrast or opposition within a statement

    oai_token-act-pair · gpt-3.5-turbo

    explicit document delimiters or end markers

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Ò
    -0.76
    thood
    -0.74
    icia
    -0.73
    imi
    -0.71
    ornings
    -0.69
    essen
    -0.69
    ESPN
    -0.66
    NB
    -0.66
    atoon
    -0.65
    .''
    -0.65
    POSITIVE LOGITS
     slightest
    1.19
     vast
    1.11
     biggest
    1.09
     majority
    1.08
     latter
    1.08
     strongest
    1.06
     greatest
    1.06
     entire
    1.06
     easiest
    1.00
     aforementioned
    1.00
    Act Density 0.295%

    No Known Activations