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

    Neuronpedia

    INDEX
    Explanations

    high or high variants

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ytale
    0.75
    立て
    0.74
    жнее
    0.72
    ゥ
    0.71
     Hooper
    0.70
     heavier
    0.68
    으로
    0.67
    ੍
    0.66
    ሐ
    0.66
    ύτε
    0.65
    POSITIVE LOGITS
     High
    3.86
    High
    3.85
     high
    3.67
    high
    3.43
     HIGH
    2.87
    HIGH
    2.85
    高
    2.64
     हाई
    2.45
     высоко
    2.35
     高
    2.34
    Act Density 0.019%

    No Known Activations