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

    Neuronpedia

    INDEX
    Explanations

    here you go

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    с
    0.61
     środ
    0.60
     şi
    0.59
    وي
    0.56
     thị
    0.55
    рез
    0.55
    ťaž
    0.55
    又は
    0.54
     daugiau
    0.54
    め
    0.54
    POSITIVE LOGITS
    na
    0.70
    ↵
    0.59
    lam
    0.58
     Signal
    0.57
     Mysteries
    0.56
     Chances
    0.54
     Limitations
    0.54
     Spoiler
    0.54
    nak
    0.53
    ↵↵↵↵
    0.52
    Act Density 0.006%

    No Known Activations