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

    Neuronpedia

    INDEX
    Explanations

    statements indicating disappointment or negative judgement

    oai_token-act-pair · gpt-3.5-turbo

    phrases expressing unfortunate circumstances or negative statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    zman
    -0.75
    orah
    -0.72
    utable
    -0.68
    æĿ
    -0.67
    ouk
    -0.64
    MH
    -0.63
    gur
    -0.63
    ×Ļ
    -0.62
    aukee
    -0.62
    inct
    -0.61
    POSITIVE LOGITS
    soever
    0.75
     happened
    0.70
     arose
    0.69
     surrounds
    0.68
    ovie
    0.67
     arises
    0.66
     happens
    0.65
     pesky
    0.65
    cher
    0.64
     occurs
    0.63
    Act Density 0.134%

    No Known Activations