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

    Neuronpedia

    INDEX
    Explanations

    numerical patterns in data, like percentages, intervals, and codes

    oai_token-act-pair · gpt-3.5-turbo

    numerical values and their associated significance in data

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    sworth
    -0.68
     Shame
    -0.64
     Cage
    -0.62
     Demand
    -0.60
     Anger
    -0.60
    stract
    -0.58
     Caption
    -0.57
    roup
    -0.57
    olas
    -0.57
     Fitzgerald
    -0.56
    POSITIVE LOGITS
    999
    0.86
    LECT
    0.81
    ãĥĩãĤ£
    0.80
    %:
    0.78
    ZX
    0.78
    eteenth
    0.74
    00
    0.73
    coni
    0.73
    ãĤ¦ãĤ¹
    0.73
    %-
    0.73
    Act Density 0.056%

    No Known Activations