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

    Neuronpedia

    INDEX
    Explanations

    Here are options

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     کشمیر
    0.71
    ार्ज
    0.68
    izz
    0.54
     Allahabad
    0.54
    ائش
    0.53
     Natl
    0.53
    偕
    0.53
     विवाद
    0.52
     Overseas
    0.52
     सकेंगे
    0.51
    POSITIVE LOGITS
     talon
    0.63
     bath
    0.60
    Garage
    0.59
     towel
    0.59
     स्पून
    0.59
     scop
    0.59
    ক্ষণ
    0.58
     metap
    0.58
     trombone
    0.58
     couper
    0.57
    Act Density 0.001%

    No Known Activations