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

    Neuronpedia

    INDEX
    Explanations

    is followed by explanation or purpose

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
        
    0.35
    etc
    0.34
    Such
    0.32
     오늘은
    0.31
     etc
    0.31
            
    0.31
     이는
    0.31
    $\
    0.30
    Which
    0.29
    或
    0.29
    POSITIVE LOGITS
     twofold
    0.85
     threefold
    0.75
     supposed
    0.57
     simply
    0.54
     probably
    0.52
     called
    0.51
     comprised
    0.50
     akin
    0.50
     actually
    0.49
     centered
    0.47
    Act Density 0.171%

    No Known Activations