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

    Neuronpedia

    INDEX
    Explanations

    compile time

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    advant
    0.36
     unqualified
    0.35
     comp
    0.34
     household
    0.34
     validity
    0.34
     bad
    0.33
     Friday
    0.33
     বর
    0.33
     outward
    0.33
     godiš
    0.33
    POSITIVE LOGITS
    कश
    0.45
    \}-
    0.43
    0.42
     imperatives
    0.42
     சிவன்
    0.42
     arsine
    0.41
    силання
    0.41
    گاهی
    0.41
    ┐
    0.41
    ீரல்
    0.41
    Act Density 0.001%

    No Known Activations