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

    Neuronpedia

    INDEX
    Explanations

    Indicating postpositions or particles

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ing
    1.22
    an
    1.21
    ة
    1.05
    Ş
    1.05
    ${
    1.02
    К
    0.98
    #
    0.98
    es
    0.96
    the
    0.96
    Μ
    0.95
    POSITIVE LOGITS
     लिए
    1.07
    ]++;
    0.93
    <0x83>
    0.89
     lavish
    0.88
     characteristic
    0.88
    いった
    0.86
    ilgan
    0.85
     घोषित
    0.84
     zealous
    0.83
     する
    0.82
    Act Density 0.011%

    No Known Activations