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

    Neuronpedia

    INDEX
    Explanations

    entities followed by parentheses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    டியா
    0.37
    gång
    0.36
    調べ
    0.36
    啶
    0.35
    Datetime
    0.35
    꺽
    0.35
     тысячи
    0.34
    的重要性
    0.34
    வருக
    0.34
    の子
    0.33
    POSITIVE LOGITS
    (
    1.64
    (-
    1.29
    複数
    0.91
    (&
    0.80
    複数の
    0.80
    (−
    0.80
    ("
    0.80
    (/
    0.77
    們
    0.76
    (,
    0.75
    Act Density 0.101%

    No Known Activations