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

    Neuronpedia

    INDEX
    Explanations

    superlatives, such as smartest, safest, simplest, darkest, deepest, and brightest

    oai_token-act-pair · gpt-3.5-turbo

    superlative adjectives describing conditions or qualities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    bara
    -0.90
    skirts
    -0.89
    mun
    -0.80
    mbuds
    -0.74
    kas
    -0.73
    bah
    -0.73
    yang
    -0.73
    ending
    -0.72
    ila
    -0.71
    lim
    -0.70
    POSITIVE LOGITS
     imaginable
    1.10
     possible
    0.98
     conceivable
    0.98
    hest
    0.86
     ever
    0.86
     Possible
    0.78
    iest
    0.78
     Worst
    0.76
     strain
    0.76
     practicable
    0.75
    Act Density 0.041%

    No Known Activations