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

    Neuronpedia

    INDEX
    Explanations

    shows you value

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     gradually
    0.42
     मसलन
    0.40
     mistakenly
    0.38
     nedenle
    0.37
     इमो
    0.37
    explained
    0.37
    aitan
    0.36
    opolis
    0.36
    Adr
    0.36
    મારા
    0.36
    POSITIVE LOGITS
    至少
    0.64
    少なくとも
    0.64
     offenbar
    0.56
    在意
    0.55
     préoccup
    0.54
    重视
    0.53
     evidently
    0.51
     गंभीरता
    0.51
     কতটা
    0.51
     atleast
    0.50
    Act Density 0.015%

    No Known Activations