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

    Neuronpedia

    INDEX
    Explanations

    adjectives with strong emotional connotations

    oai_token-act-pair · gpt-3.5-turbo

    adjectives suggesting intense or extreme qualities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    è£ħ
    -0.76
    mun
    -0.71
     teasp
    -0.70
    thur
    -0.70
    Lyn
    -0.70
    Þ
    -0.69
    ð
    -0.67
    skirts
    -0.67
     Ignore
    -0.66
    Äĩ
    -0.65
    POSITIVE LOGITS
     imaginable
    1.09
     manifestation
    0.94
     thing
    0.94
     conceivable
    0.85
     ever
    0.81
    isable
    0.80
     possible
    0.78
     exponent
    0.78
     aspects
    0.77
     manifestations
    0.76
    Act Density 0.124%

    No Known Activations