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

    Neuronpedia

    INDEX
    Explanations

    closing punctuation and commas

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     делаю
    0.47
    Separator
    0.46
     जाऊ
    0.45
    করিয়
    0.44
     मैं
    0.44
    ों
    0.43
    قر
    0.43
    াটো
    0.43
    তিক্রম
    0.42
     separado
    0.42
    POSITIVE LOGITS
    ،
    0.64
    ၊
    0.60
    ，
    0.59
    ,
    0.52
     ,
    0.49
    0.48
    ፣
    0.47
    ad
    0.45
    、「
    0.45
    蓥
    0.44
    Act Density 0.074%

    No Known Activations