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

    Neuronpedia

    INDEX
    Explanations

    disconnect from

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ↵
    0.54
    od
    0.48
     .
    0.46
    asten
    0.43
    <0xAB>
    0.43
    áng
    0.43
    ast
    0.42
    ра
    0.42
    atamente
    0.42
    ying
    0.41
    POSITIVE LOGITS
     hérit
    0.52
    츈
    0.49
    ᥣ
    0.48
    MUN
    0.48
    र्ख
    0.48
    Რ
    0.47
    asidd
    0.45
    POSIT
    0.45
    輋
    0.45
    ریک
    0.44
    Act Density 0.004%

    No Known Activations