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

    Neuronpedia

    INDEX
    Explanations

    baseline or performance level

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    り
    0.60
    นี้
    0.57
    openjdk
    0.56
    ள்ளது
    0.55
     умень
    0.54
     Contrary
    0.52
    ड़ा
    0.52
     वहीं
    0.52
    ्रेड
    0.51
     फरार
    0.51
    POSITIVE LOGITS
    I
    0.57
     সম্পদ
    0.56
     estime
    0.56
     നല്ല
    0.54
     수준
    0.54
     décisions
    0.54
     earners
    0.54
     level
    0.53
     polít
    0.53
     खरा
    0.53
    Act Density 0.007%

    No Known Activations