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

    Neuronpedia

    INDEX
    Explanations

    occurrences of the substring "fl" in words, likely indicating a focus on words related to flying or flammable themes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    elog
    -0.18
     Pied
    -0.17
    vida
    -0.16
    deaux
    -0.16
    av
    -0.16
     cater
    -0.15
    ekt
    -0.15
    placer
    -0.15
    ility
    -0.15
    inality
    -0.15
    POSITIVE LOGITS
    atter
    0.23
    ipp
    0.22
    ims
    0.21
    uster
    0.20
    ound
    0.19
    ares
    0.19
    attered
    0.18
    OUNCE
    0.17
    airs
    0.17
    ailing
    0.17
    Act Density 0.008%

    No Known Activations