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

    Neuronpedia

    INDEX
    Explanations

    designed to be capable

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     entspricht
    0.69
     saranno
    0.68
     deserves
    0.68
     still
    0.67
     será
    0.67
    จะเป็น
    0.67
     serán
    0.65
     delirium
    0.65
     serão
    0.65
     unsolved
    0.65
    POSITIVE LOGITS
     adept
    0.70
    OMET
    0.70
     capaz
    0.69
    ómetro
    0.66
     어떤
    0.63
     측정
    0.62
     capaces
    0.61
     able
    0.61
     즉
    0.60
     સત
    0.60
    Act Density 0.228%

    No Known Activations