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

    Neuronpedia

    INDEX
    Explanations

    following instructions or preferences

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    λ
    2.30
    ल
    2.19
    ي
    2.13
    л
    2.05
    й
    2.02
     hẳn
    1.92
     mưa
    1.91
    ョン
    1.77
    Giov
    1.73
    во
    1.71
    POSITIVE LOGITS
    maßen
    2.67
    蹤
    2.39
    ون
    2.14
    ك
    2.09
    ти
    1.93
    その
    1.90
    되었다
    1.88
    มีความ
    1.88
    it
    1.86
    า
    1.85
    Act Density 0.149%

    No Known Activations