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

    Neuronpedia

    INDEX
    Explanations

    writing about topics

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     gewisse
    0.75
     véritables
    0.73
    깥
    0.68
    Through
    0.65
    Während
    0.64
    executable
    0.64
    rieg
    0.63
     Through
    0.62
    บาง
    0.62
    Debug
    0.62
    POSITIVE LOGITS
     topics
    2.13
     topic
    2.07
    topics
    1.66
    topic
    1.63
     тему
    1.63
     topik
    1.58
     Topic
    1.57
     тема
    1.55
    此事
    1.51
    Topic
    1.47
    Act Density 0.424%

    No Known Activations