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

    Neuronpedia

    INDEX
    Explanations

    cover

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ery
    0.48
     Ré
    0.44
     V
    0.43
     reu
    0.39
    ero
    0.38
    éri
    0.38
    ře
    0.38
    age
    0.37
    rente
    0.37
    pero
    0.37
    POSITIVE LOGITS
    λου
    0.40
    ياة
    0.40
    cutta
    0.38
    assume
    0.38
     WHERE
    0.36
    線性
    0.36
    لون
    0.36
     gde
    0.36
     lg
    0.36
     société
    0.36
    Act Density 0.000%

    No Known Activations