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

    Neuronpedia

    INDEX
    Explanations

    superlatives or extreme degrees of comparison like "by far the most."

    oai_token-act-pair · gpt-3.5-turbo

    expressions indicating something is significantly better or greater than others

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    arnaev
    -0.73
    hao
    -0.68
    ça
    -0.67
     Sov
    -0.65
    ivated
    -0.63
    pring
    -0.63
    ppo
    -0.61
     cav
    -0.61
    itton
    -0.60
    uala
    -0.60
    POSITIVE LOGITS
     Belg
    0.68
     Unle
    0.66
     Scouting
    0.64
    -------------
    0.61
     Tradable
    0.60
     approximation
    0.60
    inois
    0.59
    è¦ļéĨĴ
    0.58
    س
    0.58
    .--
    0.57
    Act Density 0.023%

    No Known Activations