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

    Neuronpedia

    INDEX
    Explanations

    always

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     always
    1.44
     stále
    1.39
     अक्सर
    1.33
    köz
    1.30
     всегда
    1.29
    often
    1.24
    is
    1.24
     altijd
    1.22
    always
    1.22
     toujours
    1.21
    POSITIVE LOGITS
     excepting
    1.60
     bridesmaid
    1.51
     irgendwie
    1.32
     supposing
    1.25
     ending
    1.18
     darkest
    1.17
    មើ
    1.16
    ிரி
    1.13
    ज
    1.13
     пожалуйста
    1.12
    Act Density 0.048%

    No Known Activations