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

    Neuronpedia

    INDEX
    Explanations

    bullet points

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     interleaved
    0.48
     el
    0.48
     sprites
    0.46
     oscillates
    0.44
     recurs
    0.43
     lexic
    0.42
     a
    0.42
     the
    0.42
     tuple
    0.42
     regex
    0.42
    POSITIVE LOGITS
    Neben
    0.42
    <unused411>
    0.39
    Nou
    0.39
    ONU
    0.38
    ній
    0.38
    إذا
    0.38
    Chu
    0.37
    Family
    0.37
    <unused410>
    0.37
    脲
    0.36
    Act Density 2.160%

    No Known Activations