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

    Neuronpedia

    INDEX
    Explanations

    alias creation and usage

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ן
    1.68
    ні
    1.46
    ر
    1.37
    ન
    1.30
    ких
    1.26
    то
    1.23
    kého
    1.20
     motorized
    1.19
    जीशन
    1.18
    йно
    1.17
    POSITIVE LOGITS
    𝘭
    1.73
    𝘥
    1.66
    𝘨
    1.64
    𝘪
    1.52
    𝘴
    1.49
    𝘦
    1.42
    𝘰
    1.41
    𝘢
    1.40
    𝘧
    1.38
    𝘫
    1.38
    Act Density 0.001%

    No Known Activations