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

    Neuronpedia

    INDEX
    Explanations

    brain dump or paper

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ”.
    0.52
    ”。
    0.46
    %.
    0.45
    .”
    0.45
    '.
    0.42
    \".
    0.41
    ikken
    0.41
     учетом
    0.41
    ".
    0.41
    เพื่อ
    0.41
    POSITIVE LOGITS
     (?)
    0.40
     아니
    0.40
     junto
    0.38
     ]]
    0.36
     (
    0.36
    蕞
    0.36
    痈
    0.36
     During
    0.35
     (~
    0.35
     λοι
    0.35
    Act Density 0.023%

    No Known Activations