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

    Neuronpedia

    INDEX
    Explanations

    provide an answer

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    िक
    1.58
     canned
    1.52
    yn
    1.51
    ene
    1.50
    LY
    1.46
    iz
    1.45
     condu
    1.37
    ak
    1.34
    ungen
    1.33
     între
    1.31
    POSITIVE LOGITS
    т
    3.22
    t
    2.30
    檛
    1.94
    해서
    1.64
    Это
    1.64
    Ս
    1.61
    드
    1.59
    Этот
    1.58
    ﺏ
    1.56
     hề
    1.54
    Act Density 0.259%

    No Known Activations