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

    Neuronpedia

    INDEX
    Explanations

    words following a comma

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    𝐚
    1.09
    𝐞
    1.08
     collaboratively
    1.07
     collaborate
    0.98
    𝐀
    0.97
     versatile
    0.95
    𝐭
    0.94
    스와
    0.93
    𝐄
    0.92
    simmons
    0.91
    POSITIVE LOGITS
    at
    0.93
    ש
    0.92
    unused
    0.91
    れ
    0.86
    unfinished
    0.85
    у
    0.84
    child
    0.84
    η
    0.83
    al
    0.81
     doprav
    0.81
    Act Density 0.184%

    No Known Activations