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

    Neuronpedia

    INDEX
    Explanations

    Highlighting challenges

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     beetje
    0.52
    ل
    0.46
    त
    0.45
    갈
    0.44
     |\
    0.42
    ष
    0.41
    ẳ
    0.41
    节
    0.41
    可能
    0.41
    ays
    0.40
    POSITIVE LOGITS
    ir
    0.49
    ו
    0.48
    departments
    0.48
    ্রিক
    0.47
    FILTER
    0.47
    Carpenter
    0.47
    ıyla
    0.46
    inside
    0.45
    ভ্র
    0.45
    irp
    0.44
    Act Density 0.001%

    No Known Activations