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

    Neuronpedia

    INDEX
    Explanations

    program code and examples

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    𝐯
    0.59
    苷
    0.56
    ー
    0.55
    ல
    0.54
     인해
    0.54
    mär
    0.52
    воз
    0.51
    dword
    0.51
     হ্
    0.50
    𝐛
    0.50
    POSITIVE LOGITS
    ENING
    0.53
     foe
    0.50
    িল
    0.49
    য়েক
    0.46
    ies
    0.46
    گونه
    0.46
    Calc
    0.46
    पाठ
    0.45
    en
    0.44
    തന്നെ
    0.43
    Act Density 0.003%

    No Known Activations