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

    Neuronpedia

    INDEX
    Explanations

    decorators in code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     buscador
    1.13
    ismuth
    0.93
     cu
    0.91
    ada
    0.91
    azu
    0.90
    zu
    0.89
    摸
    0.88
    lo
    0.87
     thickening
    0.86
     cobre
    0.85
    POSITIVE LOGITS
    🠀
    1.35
    𝐻
    1.35
    𝑂
    1.31
     אור
    1.30
     argv
    1.30
    𝐺
    1.29
     شہریوں
    1.28
     frmt
    1.28
    tokenize
    1.27
    storms
    1.26
    Act Density 0.004%

    No Known Activations