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

    Neuronpedia

    INDEX
    Explanations

    numerical and comparative phrases or questions

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

    inquiries about how or why

    np_acts-logits-general · gemini-2.5-flash-lite

    the main finite verb or auxiliary (the predicate) of a clause/sentence.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
    DoubleQuotes
    -0.56
    脚注の使い方
    -0.50
    sizeCache
    -0.48
    참고
    -0.47
     BoxDecoration
    -0.45
    ]")]
    -0.45
     CreateTagHelper
    -0.45
     Picchu
    -0.44
    olia
    -0.44
     nakalista
    -0.43
    POSITIVE LOGITS
    oa̍t
    0.46
     nasıl
    0.45
    到底是
    0.44
     cómo
    0.41
     how
    0.41
    exactly
    0.39
    到底
    0.39
    how
    0.39
     Cómo
    0.39
     چیست
    0.38
    Act Density 0.925%

    No Known Activations