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

    Neuronpedia

    INDEX
    Explanations

    information source links

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     wins
    0.41
     decays
    0.39
     Decay
    0.36
    uyễn
    0.36
     decay
    0.35
    لية
    0.34
     difference
    0.34
    teras
    0.34
     sre
    0.34
    :>
    0.34
    POSITIVE LOGITS
     Blackmore
    0.42
    elson
    0.41
     Waterman
    0.41
    了一个
    0.37
     दोन
    0.37
    owering
    0.37
    Daemon
    0.37
    édon
    0.37
    0.37
    ondissement
    0.37
    Act Density 0.002%

    No Known Activations