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

    Neuronpedia

    INDEX
    Explanations

    question marks and periods

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    l
    1.14
    ل
    1.09
    л
    0.99
    r
    0.94
    c
    0.94
    n
    0.92
    س
    0.92
    ל
    0.90
    dır
    0.88
    园区
    0.88
    POSITIVE LOGITS
    <0x83>
    0.96
    但不
    0.93
     lowercase
    0.89
    <0x84>
    0.86
    的值
    0.84
     connotations
    0.84
     solchen
    0.82
     chord
    0.82
     vacu
    0.82
     phenylalanine
    0.81
    Act Density 0.480%

    No Known Activations