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

    Neuronpedia

    INDEX
    Explanations

    phrases related to a specific food item - "fudge."

    oai_token-act-pair · gpt-3.5-turbo

    references to the word "judge" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    spect
    -0.81
    tera
    -0.77
    ydia
    -0.71
     Stras
    -0.69
     Tec
    -0.68
    pha
    -0.67
    rises
    -0.67
    vasive
    -0.67
     XY
    -0.67
    ports
    -0.66
    POSITIVE LOGITS
    udge
    0.88
    elight
    0.78
    ules
    0.75
    atta
    0.73
    elta
    0.72
    orf
    0.71
    icket
    0.71
    edd
    0.69
    eling
    0.69
     nodd
    0.68
    Act Density 0.036%

    No Known Activations