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

    Neuronpedia

    INDEX
    Explanations

    repeated phrases that emphasize inclusivity and universal aspects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     each
    -0.68
     ciascuno
    -0.67
     każ
    -0.66
    すべて
    -0.66
     quelconque
    -0.66
     all
    -0.66
     모두
    -0.65
     ognuno
    -0.64
     algunos
    -0.63
     Each
    -0.62
    POSITIVE LOGITS
    together
    0.87
    though
    0.84
    uding
    0.83
     kinds
    0.80
     sorts
    0.79
     three
    0.78
    ograft
    0.76
    rounder
    0.75
     else
    0.75
    ergies
    0.74
    Act Density 0.223%

    No Known Activations