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

    Neuronpedia

    INDEX
    Explanations

    multi-script tokens followed by common suffixes/related words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    et
    0.90
    an
    0.86
    il
    0.76
    o
    0.71
    ir
    0.71
    ed
    0.70
    a
    0.70
    ia
    0.63
    at
    0.61
    ab
    0.60
    POSITIVE LOGITS
    会
    0.71
    ने
    0.60
    在
    0.60
    да
    0.60
    も
    0.59
    ᱖
    0.59
    ﺔ
    0.57
    신
    0.55
    ﻘ
    0.54
    회
    0.54
    Act Density 0.103%

    No Known Activations