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

    Neuronpedia

    INDEX
    Explanations

    right and assignment

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    чні
    0.47
    𝟰
    0.42
    𝗳
    0.42
     아름
    0.40
     మ‌
    0.40
    批判
    0.39
     ಮತ್ತ
    0.39
    чин
    0.39
    িয়াছেন
    0.38
    чках
    0.38
    POSITIVE LOGITS
     counterpart
    0.54
     equally
    0.49
     both
    0.42
     respectively
    0.41
     Both
    0.40
     similarly
    0.40
     counterparts
    0.38
    "
    0.38
     correspondingly
    0.38
    ".
    0.38
    Act Density 0.181%

    No Known Activations