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

    Neuronpedia

    INDEX
    Explanations

    someone actions or descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    an
    1.75
    ம்
    1.58
    y
    1.52
    a
    1.30
    ا
    1.26
    ură
    1.20
    iere
    1.19
    ações
    1.18
    ală
    1.17
     logrado
    1.16
    POSITIVE LOGITS
     else
    1.87
     із
    1.35
    else
    1.33
    ⁢</
    1.22
     щодо
    1.20
    ဦ
    1.20
     факт
    1.20
    ੁ
    1.20
     montée
    1.16
    är
    1.15
    Act Density 0.282%

    No Known Activations