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

    Neuronpedia

    INDEX
    Explanations

    contrasting proprietary vs open

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     was
    1.04
     renormal
    1.02
     and
    0.96
     reorgan
    0.96
     {
    0.95
     is
    0.89
    ва
    0.88
     emergence
    0.88
     είχε
    0.87
     he
    0.86
    POSITIVE LOGITS
    ir
    1.33
    unlike
    1.27
    Unlike
    1.26
    im
    1.20
    os
    1.19
    uts
    1.17
    یت
    1.02
    я
    1.02
    uk
    0.98
    il
    0.97
    Act Density 0.004%

    No Known Activations