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

    Neuronpedia

    INDEX
    Explanations

    phrases related to the purpose or intent of an action

    oai_token-act-pair · gpt-3.5-turbo

    references to various usages or purposes for actions and materials

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    BUG
    -0.71
    Ott
    -0.67
    esc
    -0.67
    STRUCT
    -0.66
    oyer
    -0.66
    tro
    -0.64
    Lat
    -0.63
    ãģį
    -0.63
    ãĥ¼ãĥĨãĤ£
    -0.63
    Walk
    -0.62
    POSITIVE LOGITS
    afe
    0.82
    pring
    0.77
     involving
    0.77
    chool
    0.74
    heet
    0.74
     stemming
    0.71
     pertaining
    0.69
    atics
    0.69
     whatsoever
    0.67
     relating
    0.67
    Act Density 0.229%

    No Known Activations