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

    Neuronpedia

    INDEX
    Explanations

    items or entities

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     железнодоро
    0.44
    人们
    0.43
     자동차
    0.39
     pharmacological
    0.38
    avimo
    0.38
    laboratory
    0.37
    âmara
    0.37
     특별
    0.37
    房地产
    0.37
     terapé
    0.36
    POSITIVE LOGITS
     ones
    0.85
     item
    0.82
     thing
    0.72
     piece
    0.71
     guy
    0.61
     spot
    0.58
     option
    0.58
     utterance
    0.57
     instance
    0.56
     outfit
    0.55
    Act Density 2.278%

    No Known Activations