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

    Neuronpedia

    INDEX
    Explanations

    end of phrases followed by punctuation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Whenever
    0.47
     jeżeli
    0.46
     যদি
    0.45
     hvis
    0.44
     যখন
    0.42
     Jeżeli
    0.42
     Porque
    0.40
     रहेको
    0.39
    ując
    0.39
     Owing
    0.38
    POSITIVE LOGITS
    ،
    0.69
     there
    0.68
     it
    0.67
     thì
    0.62
    ၊
    0.59
    ，
    0.58
    ,
    0.50
    [,]
    0.50
     එය
    0.48
    there
    0.48
    Act Density 0.009%

    No Known Activations