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

    Neuronpedia

    INDEX
    Explanations

    OpenAI, learning, velocity

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    chek
    1.29
    *}\
    1.18
    𝑰
    1.17
     idade
    1.12
    𝗼
    1.10
    }(
    1.10
    𝟬
    1.09
     Appart
    1.09
    mittag
    1.08
    ise
    1.08
    POSITIVE LOGITS
    ের
    1.39
    менение
    1.29
     treacher
    1.24
     tubercle
    1.23
    нің
    1.17
     Balmer
    1.15
    RAW
    1.14
     gjennom
    1.14
    誣
    1.14
    smoothed
    1.10
    Act Density 0.000%

    No Known Activations