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

    Neuronpedia

    INDEX
    Explanations

    occurrences of a specific sequence of characters that seem to pertain to names or titles

    oai_token-act-pair · gpt-3.5-turbo

    the names of specific characters or entities, particularly those related to "Cha" and "Sha."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ragon
    -0.75
    tons
    -0.66
    worthiness
    -0.62
    MAS
    -0.62
    lings
    -0.61
    wise
    -0.61
    mask
    -0.61
    scale
    -0.60
    t
    -0.60
    pants
    -0.60
    POSITIVE LOGITS
    plin
    1.34
    vern
    1.14
    uble
    1.08
    pling
    1.06
    vel
    1.03
    pless
    1.03
    plain
    1.02
    usal
    1.01
    ï
    1.00
    umann
    1.00
    Act Density 0.073%

    No Known Activations