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

    Neuronpedia

    INDEX
    Explanations

    defining objectives

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     neoplas
    0.44
     idE
    0.40
     தங்கள்
    0.39
    ργαν
    0.39
    国内外
    0.39
     potranno
    0.39
     தேர்ந்த
    0.38
     ehemaligen
    0.38
    claiming
    0.38
    gcj
    0.38
    POSITIVE LOGITS
     często
    0.55
     often
    0.53
     quando
    0.52
     Often
    0.50
     často
    0.50
     When
    0.50
    往往
    0.50
     unspoken
    0.50
     interrupt
    0.50
     cuando
    0.49
    Act Density 0.341%

    No Known Activations