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

    Neuronpedia

    INDEX
    Explanations

    right now, compared to

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     정말
    0.44
     într
    0.40
     davon
    0.39
     demek
    0.39
     definitely
    0.39
    จริงๆ
    0.38
     beware
    0.38
     περισσότε
    0.37
     naar
    0.37
    гти
    0.36
    POSITIVE LOGITS
    their
    0.59
     their
    0.58
     их
    0.51
     Their
    0.50
    Their
    0.48
     তাদের
    0.47
     ಅವರ
    0.47
     kanilang
    0.45
    他们的
    0.44
    coal
    0.44
    Act Density 0.124%

    No Known Activations