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

    Neuronpedia

    INDEX
    Explanations

    math problems

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     любой
    -0.09
     будь
    -0.08
     consider
    -0.08
    =L
    -0.08
    meaning
    -0.08
     любое
    -0.08
     каждом
    -0.08
    任何
    -0.08
     buen
    -0.08
     chaque
    -0.08
    POSITIVE LOGITS
     aparentemente
    0.12
     apparently
    0.11
     schein
    0.11
     offenbar
    0.11
    -ish
    0.10
     Seems
    0.10
     Apparently
    0.09
     específicos
    0.09
     seemingly
    0.09
    (?)
    0.09
    Act Density 0.066%

    No Known Activations