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

    Neuronpedia

    INDEX
    Explanations

    barely a whisper or audible

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ັນ
    0.48
     ONLY
    0.47
    อ
    0.46
     relatively
    0.44
     fairly
    0.43
    比較的
    0.43
     die
    0.43
     comparatively
    0.43
     only
    0.42
     bland
    0.42
    POSITIVE LOGITS
     barely
    0.58
    above
    0.52
    高于
    0.51
    上記
    0.48
    Above
    0.48
    合格
    0.47
     acima
    0.46
     Above
    0.46
    icelli
    0.46
     above
    0.45
    Act Density 0.008%

    No Known Activations