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

    Neuronpedia

    INDEX
    Explanations

    words and phrases related to alternatives or substitutions

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to substitutes or alternatives

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    oji
    -0.80
    iland
    -0.74
    DEN
    -0.68
    ahime
    -0.67
    behind
    -0.65
    ipl
    -0.63
    umbered
    -0.63
    uck
    -0.62
    essee
    -0.62
    ggle
    -0.62
    POSITIVE LOGITS
     existing
    0.69
     succeeding
    0.65
     rotating
    0.63
     pree
    0.62
    ctica
    0.62
     subsistence
    0.62
     usual
    0.62
     satisfying
    0.61
     Micha
    0.60
     foregoing
    0.60
    Act Density 0.243%

    No Known Activations