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

    Neuronpedia

    INDEX
    Explanations

    instructions, steps, solutions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     всего
    1.30
     fazla
    1.20
     gode
    1.17
    諧
    1.17
     peut
    1.17
     jeste
    1.16
     iš
    1.16
     toekom
    1.15
     '}
    1.15
    ﺩ
    1.12
    POSITIVE LOGITS
     jornada
    0.97
    pads
    0.96
    oč
    0.96
    Τα
    0.94
     comple
    0.94
    如下图
    0.93
     জাক
    0.93
    ✦
    0.91
    vast
    0.90
    multip
    0.89
    Act Density 0.000%

    No Known Activations