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

    Neuronpedia

    INDEX
    Explanations

    alternatives, but, languages

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

    capitalized proper nouns and acronyms in technical contexts, often adjacent to punctuation like commas or apostrophes.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
    गिंग
    0.39
     CompoundButton
    0.39
    nameof
    0.38
    quite
    0.38
    Quite
    0.38
    ListOf
    0.37
    某个
    0.36
    adım
    0.36
    navigation
    0.35
    经常
    0.35
    POSITIVE LOGITS
    以外の
    0.59
     Alternatives
    0.55
     tradicional
    0.51
    以外
    0.50
     But
    0.49
     tradizionale
    0.49
     alternatives
    0.48
     além
    0.48
     but
    0.47
     tradicionales
    0.47
    Act Density 0.371%

    No Known Activations