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

    Neuronpedia

    INDEX
    Explanations

    AI assistant refusing harmful requests

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     flax
    0.46
     cacao
    0.44
    合作
    0.39
    আনু
    0.38
    elv
    0.38
    昏
    0.38
     sacar
    0.38
    elina
    0.37
     dast
    0.37
     levi
    0.37
    POSITIVE LOGITS
    ﻀ
    0.41
    getReference
    0.36
     అంత
    0.35
    Insert
    0.34
     விளங்க
    0.34
    गौर
    0.34
    鳳
    0.33
    のと
    0.33
     വഹ
    0.33
    홀
    0.33
    Act Density 0.010%

    No Known Activations