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

    Neuronpedia

    INDEX
    Explanations

    access, digest, memorizing, worth

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    री
    0.56
    m
    0.53
     occasional
    0.51
    म
    0.50
    ਮ
    0.48
     colored
    0.47
    地方
    0.47
     wolf
    0.46
    嵐
    0.46
    Wolf
    0.45
    POSITIVE LOGITS
     этому
    0.54
    ümer
    0.52
     فيلم
    0.52
    蛜
    0.52
    are
    0.51
     чему
    0.49
     tük
    0.48
    èvement
    0.47
    नाइटेड
    0.47
    蒉
    0.47
    Act Density 0.000%

    No Known Activations