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

    Neuronpedia

    INDEX
    Explanations

    ethical considerations and safety guidelines

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    uq
    0.44
     Its
    0.43
     yeter
    0.43
     GAM
    0.42
    ஸ்
    0.40
    alow
    0.40
     longer
    0.40
     queens
    0.40
    GAM
    0.40
     faker
    0.40
    POSITIVE LOGITS
    ienes
    0.47
    ",
    0.46
    ons
    0.45
    rées
    0.45
    ancipation
    0.44
    0.43
     próxim
    0.42
    ines
    0.42
    olo
    0.41
    COLLATION
    0.41
    Act Density 0.002%

    No Known Activations