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

    Neuronpedia

    INDEX
    Explanations

    still ongoing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    了吧
    0.74
    一看
    0.71
     하게
    0.69
     increasing
    0.67
    नात
    0.67
     renewed
    0.66
     increasingly
    0.66
     useless
    0.65
    ivert
    0.65
     تصير
    0.65
    POSITIVE LOGITS
     still
    3.45
    still
    3.13
     아직
    3.12
     Still
    3.02
    Still
    2.99
     masih
    2.91
     هنوز
    2.90
    まだ
    2.89
     todavía
    2.85
     ainda
    2.75
    Act Density 0.203%

    No Known Activations