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

    Neuronpedia

    INDEX
    Explanations

    the word "is" at the beginning of a sentence

    oai_token-act-pair · gpt-3.5-turbo

    repeated phrases or structures that start with "this is."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    isms
    -0.67
    luaj
    -0.66
    nesses
    -0.66
    selves
    -0.65
     interven
    -0.64
    forts
    -0.64
    cknow
    -0.61
    occup
    -0.61
    ellig
    -0.60
    Parameters
    -0.60
    POSITIVE LOGITS
     my
    0.94
     definitely
    0.84
     why
    0.84
     gonna
    0.83
     how
    0.83
     NOT
    0.78
     probably
    0.78
     what
    0.77
     an
    0.75
     another
    0.73
    Act Density 0.102%

    No Known Activations