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

    Neuronpedia

    INDEX
    Explanations

    references and C++ code constructs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    0.69
    do
    0.66
    that
    0.65
    ves
    0.63
    ᱐
    0.63
    sten
    0.63
    gat
    0.62
    ta
    0.61
    án
    0.61
    ten
    0.60
    POSITIVE LOGITS
    س
    0.74
     ambience
    0.69
    ار
    0.69
    鹘
    0.69
    の
    0.66
     Bibliothèque
    0.66
    ل
    0.65
     nymphs
    0.64
     planners
    0.64
     peculiarities
    0.64
    Act Density 0.002%

    No Known Activations