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

    Neuronpedia

    INDEX
    Explanations

    click, steps, trends

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     olvidar
    0.42
    辦法
    0.42
    siye
    0.40
    quando
    0.38
     বাসিন্দা
    0.38
     Invis
    0.37
    vis
    0.37
     obras
    0.37
     तभी
    0.37
     deterg
    0.37
    POSITIVE LOGITS
    姚
    0.39
     steered
    0.38
    нде
    0.35
    և
    0.35
     Bü
    0.35
     කල
    0.33
     Wieder
    0.33
     neue
    0.33
     ξεκ
    0.33
    தைத்
    0.33
    Act Density 0.000%

    No Known Activations