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

    Neuronpedia

    INDEX
    Explanations

    LaTeX and code structuring

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Locations
    0.43
     locations
    0.42
     margin
    0.41
     Det
    0.41
    margin
    0.40
    locations
    0.40
     det
    0.39
    Det
    0.38
     stalls
    0.38
     robust
    0.37
    POSITIVE LOGITS
    ссер
    0.45
     PLANE
    0.43
     зу
    0.39
     محسوس
    0.39
     LAYER
    0.39
    ક્ત
    0.39
     chegada
    0.38
    경기도
    0.38
    ぞ
    0.38
     MAINTENANCE
    0.37
    Act Density 0.001%

    No Known Activations