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

    Neuronpedia

    INDEX
    Explanations

    increase in

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    𝐫
    1.49
    ला
    1.37
    <0xB9>
    1.32
    م
    1.26
    <0xA5>
    1.23
     azules
    1.23
    𝐡
    1.22
    𝘀
    1.21
    м
    1.20
    𝐥
    1.19
    POSITIVE LOGITS
    Њ
    1.37
    able
    1.19
    вано
    1.18
     SKSpriteNode
    1.15
    ى
    1.12
     sott
    1.09
     barometer
    1.09
    います
    1.06
     termina
    1.05
    ю
    1.05
    Act Density 0.002%

    No Known Activations