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

    Neuronpedia

    INDEX
    Explanations

    computational and linguistic tasks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     favourites
    0.47
    ing
    0.46
    h
    0.46
    ;
    0.45
     글로벌
    0.44
     еще
    0.43
    était
    0.42
    archical
    0.42
    est
    0.41
     이미
    0.41
    POSITIVE LOGITS
     thereupon
    0.54
     বৎসর
    0.50
    作為
    0.50
     whereupon
    0.47
     ως
    0.46
     Olson
    0.46
    ŭ
    0.44
     entsprechend
    0.44
    رځ
    0.43
    ril
    0.43
    Act Density 0.001%

    No Known Activations