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

    Neuronpedia

    INDEX
    Explanations

    list item separators

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    是我们
    0.30
     insgesamt
    0.29
     Owing
    0.29
     następnie
    0.29
     Thereby
    0.29
     przede
    0.29
     sehingga
    0.28
     जिससे
    0.28
     honom
    0.27
     Significantly
    0.27
    POSITIVE LOGITS
    ,
    0.61
    ،
    0.57
    ，
    0.52
     there
    0.50
     things
    0.39
    ၊
    0.39
    、
    0.38
     অবশ্য
    0.38
    there
    0.37
     however
    0.37
    Act Density 0.140%

    No Known Activations