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

    Neuronpedia

    INDEX
    Explanations

    positive adjectives, especially related to fantasy or exceptional imagery

    oai_token-act-pair · gpt-3.5-turbo

    references to fantastical elements and specific media titles

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Kings
    -0.78
    keeper
    -0.68
    ership
    -0.66
    Combat
    -0.65
    Marcus
    -0.64
    Rush
    -0.64
    den
    -0.64
    Minecraft
    -0.63
    ding
    -0.63
    guard
    -0.60
    POSITIVE LOGITS
    astical
    1.08
     Fant
    0.96
     EF
    0.84
    imes
    0.82
    idays
    0.79
    antom
    0.77
    thora
    0.75
     FANT
    0.74
    onial
    0.73
    onies
    0.73
    Act Density 0.027%

    No Known Activations