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

    Neuronpedia

    INDEX
    Explanations

    long-range dependencies in sequences

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    extré
    0.41
     ionized
    0.40
     regl
    0.39
     excluded
    0.39
    ity
    0.38
     restre
    0.38
    strup
    0.38
    ?-
    0.38
    itore
    0.38
    illons
    0.38
    POSITIVE LOGITS
     harina
    0.46
    荣
    0.43
     ойной
    0.43
     knives
    0.42
     giz
    0.41
    י
    0.41
     spoons
    0.39
    🥄
    0.39
     जीना
    0.39
    榮
    0.39
    Act Density 0.002%

    No Known Activations