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

    Neuronpedia

    INDEX
    Explanations

    processing complexity and speed

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     acus
    0.43
     accusations
    0.41
     Hormone
    0.39
    чина
    0.38
     üle
    0.38
     Bf
    0.38
     hạt
    0.38
     Greenfield
    0.37
    чы
    0.36
    酢
    0.36
    POSITIVE LOGITS
    原始内容
    0.41
     served
    0.40
     winding
    0.39
    served
    0.39
    湍
    0.39
    🌋
    0.38
     gravitation
    0.38
    postion
    0.37
    leece
    0.37
     ஏற
    0.37
    Act Density 0.000%

    No Known Activations