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

    Neuronpedia

    INDEX
    Explanations

    intensified or superlative adjectives

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    "very" followed by descriptive word

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    tanleria
    -0.82
    Rujuakan
    -0.78
     kaarangay
    -0.75
     informée
    -0.74
     للاسماء
    -0.74
    verwijspagina
    -0.71
     الرياضيه
    -0.69
     Italijanski
    -0.68
     насељу
    -0.65
     мәкал
    -0.64
    POSITIVE LOGITS
     super
    0.41
     súper
    0.35
     very
    0.35
     famous
    0.34
     too
    0.34
     necessary
    0.33
     bete
    0.31
    function
    0.29
     function
    0.29
    Called
    0.28
    Act Density 0.015%

    No Known Activations