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

    Neuronpedia

    INDEX
    Explanations

    references to hotel names, specifically those ending with "Carlton."

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the substring "itz."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    åŃ
    -0.66
     doub
    -0.64
     unheard
    -0.64
    selves
    -0.63
     Malays
    -0.63
     oppressed
    -0.63
     occas
    -0.62
     readable
    -0.61
    ieties
    -0.61
    ised
    -0.59
    POSITIVE LOGITS
    gerald
    1.81
    sch
    1.18
    patrick
    1.18
    enger
    1.13
    heimer
    1.12
    enegger
    1.10
    ky
    0.99
    enberg
    0.98
    arella
    0.96
    mann
    0.96
    Act Density 0.028%

    No Known Activations