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

    Neuronpedia

    INDEX
    Explanations

    introducing a new idea,

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     নইলে
    0.33
     সুতরাং
    0.30
     honom
    0.29
    但是我
    0.29
     जिससे
    0.29
    所以我
    0.29
     Hence
    0.28
     然後
    0.28
     lesquels
    0.27
     mutta
    0.26
    POSITIVE LOGITS
    ،
    0.72
    ,
    0.67
    ，
    0.65
    ၊
    0.55
    、
    0.49
    ፣
    0.47
     there
    0.44
     it
    0.42
    呢
    0.41
     we
    0.39
    Act Density 0.062%

    No Known Activations