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

    Neuronpedia

    INDEX
    Explanations

    ash, bash, mesh, ashka

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     fresh
    1.08
     robust
    0.82
    ened
    0.80
    ❰
    0.78
    рио
    0.78
     shady
    0.77
    echo
    0.76
     freshman
    0.76
    essero
    0.76
    eous
    0.76
    POSITIVE LOGITS
    ׁ
    1.54
    s
    1.28
    midt
    1.27
    ের
    1.25
    ׂ
    1.19
    awn
    1.16
    apixel
    1.13
    ים
    1.09
    يء
    1.09
    еры
    1.08
    Act Density 0.109%

    No Known Activations