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

    Neuronpedia

    INDEX
    Explanations

    mail related phrases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ı
    2.36
    í
    2.34
    ς
    2.19
    s
    2.14
    い
    2.13
    而且
    2.03
    fen
    2.00
    jenigen
    1.91
    ü
    1.89
    י
    1.87
    POSITIVE LOGITS
    ن
    2.33
    끔
    2.16
    ную
    2.00
    りに
    1.88
     sturdy
    1.87
     overcame
    1.85
     overseeing
    1.82
    нский
    1.80
     उन्नति
    1.80
    ない
    1.79
    Act Density 0.044%

    No Known Activations