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

    Neuronpedia

    INDEX
    Explanations

    discussion of policy and value

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ك
    0.50
    uerpo
    0.47
     گ
    0.45
     دي
    0.44
     الب
    0.44
     ب
    0.43
     pate
    0.43
     الك
    0.43
    ([-
    0.43
     الس
    0.43
    POSITIVE LOGITS
    अधिकांश
    0.44
    🟦
    0.41
    이제
    0.41
    🛸
    0.40
    <unused2150>
    0.40
    🩶
    0.40
    극장
    0.39
    🤎
    0.39
    💟
    0.39
    🫶
    0.39
    Act Density 0.000%

    No Known Activations