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

    Neuronpedia

    INDEX
    Explanations

    numerical percentages or quantities preceded by a numerical value

    oai_token-act-pair · gpt-3.5-turbo

    numerical values associated with specific statistics and percentages

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ettings
    -0.72
    venge
    -0.68
    Fram
    -0.68
    medi
    -0.68
    vati
    -0.65
    folk
    -0.62
    hunt
    -0.60
     judgement
    -0.60
    actic
    -0.58
    yright
    -0.57
    POSITIVE LOGITS
    %
    1.12
    %-
    1.09
     percent
    1.06
    %"
    1.05
    %.
    1.01
     Shades
    0.97
     Minutes
    0.97
    %;
    0.94
    %,
    0.92
    %:
    0.91
    Act Density 0.165%

    No Known Activations