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

    Neuronpedia

    INDEX
    Explanations

    mean squared error

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    expects
    0.80
     cauza
    0.80
    ्राफी
    0.79
     annuelle
    0.78
     вследствие
    0.78
     parfaitement
    0.76
     dvara
    0.76
     ezingu
    0.75
     aisément
    0.75
    粱
    0.75
    POSITIVE LOGITS
    ె
    0.79
    ä
    0.77
    c
    0.72
     الحس
    0.68
     remnants
    0.60
    вая
    0.59
    மாக
    0.59
     h
    0.59
    ılık
    0.58
    S
    0.58
    Act Density 0.000%

    No Known Activations