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

    Neuronpedia

    INDEX
    Explanations

    U

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     plz
    0.50
     symbols
    0.41
    dí
    0.41
     tienda
    0.40
     ponte
    0.40
     bú
    0.40
     símbolos
    0.40
     simbol
    0.40
     recebeu
    0.40
     रिवाज
    0.40
    POSITIVE LOGITS
     Nus
    0.37
    𝗞
    0.37
    multiply
    0.36
    fassung
    0.36
     колле
    0.36
    singular
    0.36
    0.35
    getClass
    0.35
     Tep
    0.34
    穎
    0.34
    Act Density 0.002%

    No Known Activations