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

    Neuronpedia

    INDEX
    Explanations

    (1-5, 1 being)

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    代替
    0.39
     alternative
    0.38
     başka
    0.37
     oluştur
    0.37
    ндай
    0.37
    ই
    0.36
     alternatif
    0.35
     cilvē
    0.35
    पचारिक
    0.33
     čovjek
    0.33
    POSITIVE LOGITS
    其中
    0.74
    のうち
    0.73
     其中
    0.71
     большинство
    0.70
     közül
    0.66
     whereof
    0.64
     некоторые
    0.59
    पैकी
    0.59
     অধিকাংশই
    0.58
     některé
    0.58
    Act Density 0.133%

    No Known Activations