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

    Neuronpedia

    INDEX
    Explanations

    instances where the verb 'will' is used along with related terms like 'copy', 'title', or 'raising'

    oai_token-act-pair · gpt-3.5-turbo

    references to actions and outcomes relating to titles, names, and attributes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Dixon
    -0.76
     Spectre
    -0.71
     Moss
    -0.62
     Sinclair
    -0.62
     Mull
    -0.62
    PDATE
    -0.61
     Gould
    -0.61
    LINE
    -0.60
    IDA
    -0.60
    Sullivan
    -0.60
    POSITIVE LOGITS
     be
    0.84
    ãĥĻ
    0.78
    being
    0.77
     is
    0.74
    amar
    0.73
     being
    0.72
    ãĤ¹
    0.72
    ãĥĵ
    0.68
    be
    0.67
    igate
    0.66
    Act Density 0.237%

    No Known Activations