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

    Neuronpedia

    INDEX
    Explanations

    long distance/range

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    laze
    -0.77
    锢
    -0.73
     equilibrium
    -0.73
    吭
    -0.72
     тю
    -0.71
     IPT
    -0.71
    覆
    -0.71
     Ochoa
    -0.71
    bera
    -0.70
     Zep
    -0.70
    POSITIVE LOGITS
     long
    2.44
    long
    1.94
    Long
    1.91
     Long
    1.83
    LONG
    1.57
    长
    1.54
    長
    1.48
     LONG
    1.41
     длин
    1.41
     dlou
    1.40
    Act Density 0.021%

    No Known Activations