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

    Neuronpedia

    INDEX
    Explanations

    difficulty levels and categories

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    that
    0.67
     totiž
    0.61
    the
    0.59
    this
    0.56
    将
    0.54
    在这里
    0.53
    它们的
    0.52
    api
    0.51
    That
    0.51
    它们
    0.51
    POSITIVE LOGITS
     dgn
    0.84
     עם
    0.82
     др
    0.74
     근데
    0.72
     กับ
    0.68
     lakini
    0.66
     อาจ
    0.66
     với
    0.65
     nhưng
    0.64
     illetve
    0.64
    Act Density 0.049%

    No Known Activations