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

    Neuronpedia

    INDEX
    Explanations

    Who is who

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     existem
    0.45
     infinitely
    0.40
     multipurpose
    0.39
    有两种
    0.38
    regarding
    0.38
    几种
    0.38
     imperfect
    0.37
    شوند
    0.37
     esistono
    0.37
    esistenza
    0.37
    POSITIVE LOGITS
     Who
    0.74
     notables
    0.73
    Who
    0.68
     roster
    0.66
     impressive
    0.63
     enviable
    0.62
     Кто
    0.59
    錚
    0.59
     кто
    0.59
    who
    0.58
    Act Density 0.054%

    No Known Activations