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

    Neuronpedia

    INDEX
    Explanations

    detecting data types

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ים
    0.50
    etition
    0.48
    ारे
    0.47
    ילות
    0.47
    ারের
    0.47
    ля
    0.46
    ิ่น
    0.44
    s
    0.44
    outs
    0.44
    ing
    0.42
    POSITIVE LOGITS
     ਫ
    0.52
     corret
    0.52
     هڪ
    0.50
     weten
    0.49
     connu
    0.49
    知道
    0.48
    COOL
    0.47
    ᕝ
    0.47
    秏
    0.46
     知
    0.46
    Act Density 0.000%

    No Known Activations