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

    Neuronpedia

    INDEX
    Explanations

    entities and groups

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     использует
    0.82
     sometimes
    0.81
     தெரிவித்த
    0.80
     parfois
    0.77
    sometimes
    0.77
    requently
    0.75
     maintains
    0.75
     may
    0.75
     strives
    0.75
     ofrece
    0.75
    POSITIVE LOGITS
     companies
    1.52
     corporations
    1.36
     الناس
    1.28
     Companies
    1.23
     people
    1.21
     insurers
    1.19
     governments
    1.18
     ludzie
    1.16
     advertisers
    1.15
    companies
    1.15
    Act Density 0.041%

    No Known Activations