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

    Neuronpedia

    INDEX
    Explanations

    HTML tags and structural elements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ↵↵
    0.86
    ，
    0.77
     of
    0.75
     (
    0.60
     that
    0.60
        
    0.59
     這
    0.57
    ↵↵↵
    0.56
                    
    0.56
          
    0.55
    POSITIVE LOGITS
    ي
    0.82
    و
    0.77
    ת
    0.77
    us
    0.76
    ри
    0.68
    га
    0.67
    もら
    0.67
    usd
    0.64
    um
    0.64
    𝗱
    0.63
    Act Density 0.023%

    No Known Activations