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

    Neuronpedia

    INDEX
    Explanations

    Turkish suffixes and Cyrillic 'chi'/'ci'

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    hosts
    0.45
    들아
    0.43
    individuals
    0.40
     vik
    0.40
     COMPENSATION
    0.39
     families
    0.39
     країн
    0.38
    들과
    0.37
     Communists
    0.37
    Compensation
    0.37
    POSITIVE LOGITS
    چی
    1.02
    ci
    0.93
    чи
    0.89
    chi
    0.82
    CI
    0.81
    çi
    0.80
    cı
    0.79
    çı
    0.75
    cil
    0.75
    чия
    0.70
    Act Density 0.001%

    No Known Activations