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

    Neuronpedia

    INDEX
    Explanations

    androidx imports android

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ה
    2.15
    ્સ
    1.97
     manqu
    1.93
    akaranam
    1.90
    imensional
    1.87
     enchanted
    1.83
     öğrenc
    1.83
     ‬
    1.82
     surmounted
    1.81
    ન
    1.81
    POSITIVE LOGITS
    <bos>
    1.52
    we
    1.41
     thunder
    1.39
    prac
    1.39
    人
    1.35
     stiff
    1.34
    バター
    1.32
    mate
    1.31
    ler
    1.31
     lofty
    1.29
    Act Density 0.000%

    No Known Activations