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

    Neuronpedia

    INDEX
    Explanations

    words or scripts

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    менование
    0.46
     डाइ
    0.41
     edeb
    0.39
    憚
    0.37
    ae
    0.37
    ayv
    0.37
    гре
    0.36
    েন্টস
    0.36
    treat
    0.36
    惮
    0.35
    POSITIVE LOGITS
     ure
    0.51
    и
    0.45
    enc
    0.44
     تشوف
    0.44
     Enc
    0.43
     ঔ
    0.42
    uffy
    0.40
    ures
    0.40
    estra
    0.38
    ина
    0.37
    Act Density 0.001%

    No Known Activations