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

    Neuronpedia

    INDEX
    Explanations

    forecasting and model performance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ᱘
    0.49
     yǒu
    0.47
     має
    0.46
     zamów
    0.45
    نساء
    0.44
    ÈRE
    0.43
    ኗ
    0.42
     ಪೊ
    0.42
    チナ
    0.42
     島
    0.42
    POSITIVE LOGITS
    that
    0.51
    j
    0.49
    ز
    0.49
    م
    0.48
    c
    0.48
    jing
    0.47
    jm
    0.45
    jada
    0.45
     எடுத்துக்
    0.44
    غت
    0.44
    Act Density 0.002%

    No Known Activations