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

    Neuronpedia

    INDEX
    Explanations

    character types

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    COMANDA
    0.57
     zaměst
    0.55
     metabolismo
    0.54
    投資
    0.54
    Gosudarstvennyj
    0.53
     gobierno
    0.52
     podnik
    0.52
    ARAJYA
    0.52
    規模
    0.51
     organisasi
    0.51
    POSITIVE LOGITS
     characters
    0.83
    字符
    0.80
     punctuation
    0.77
     ASCII
    0.71
     chars
    0.71
     Unicode
    0.70
     Characters
    0.70
    chars
    0.68
    characters
    0.66
    Chars
    0.66
    Act Density 0.466%

    No Known Activations