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

    Neuronpedia

    INDEX
    Explanations

    sentence start after period

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    不仅
    0.50
    致力于
    0.50
    Jeśli
    0.50
    Você
    0.47
    姏
    0.45
    Với
    0.45
    🛍
    0.45
     unsurprisingly
    0.45
    <unused2049>
    0.45
    Bạn
    0.44
    POSITIVE LOGITS
     was
    0.51
     tigers
    0.50
     (
    0.46
     two
    0.46
     m
    0.45
     ty
    0.45
     soldiers
    0.45
     "
    0.45
    0.45
     killers
    0.44
    Act Density 0.015%

    No Known Activations