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

    Neuronpedia

    INDEX
    Explanations

    overriding methods in code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ième
    1.85
    นะคะ
    1.66
    ä
    1.63
    ्ञ
    1.62
    ст
    1.59
    ты
    1.56
    기
    1.54
    ја
    1.49
    ía
    1.48
    ú
    1.47
    POSITIVE LOGITS
    s
    1.68
    ς
    1.61
    ﻥ
    1.48
     dotycz
    1.44
    Name
    1.40
    ن
    1.40
    an
    1.38
    кові
    1.37
     Perg
    1.35
    en
    1.34
    Act Density 0.000%

    No Known Activations