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

    Neuronpedia

    INDEX
    Explanations

    governs your use

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     iodide
    0.43
     animación
    0.43
     greenery
    0.39
     tassels
    0.38
     xyl
    0.37
     dandelion
    0.36
     fá
    0.36
     cinta
    0.36
     vaiht
    0.36
    嘱
    0.36
    POSITIVE LOGITS
    е
    0.49
    et
    0.45
    legal
    0.41
    а
    0.41
    uns
    0.40
    see
    0.39
    person
    0.38
    ies
    0.38
    id
    0.38
    pr
    0.38
    Act Density 0.001%

    No Known Activations