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

    Neuronpedia

    INDEX
    Explanations

    GitHub links and URLs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ,
    1.30
    .
    1.29
    luor
    0.95
    -
    0.90
     (
    0.90
     only
    0.89
     assumes
    0.89
     server
    0.86
     cloud
    0.84
     =
    0.83
    POSITIVE LOGITS
    𝔹
    1.50
    𝙽
    1.49
    𝚃
    1.48
    𝚁
    1.48
    𝑯
    1.46
    𝙲
    1.44
    𝗖
    1.44
    ZU
    1.42
    Hrs
    1.42
    Вели
    1.39
    Act Density 0.098%

    No Known Activations