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

    Neuronpedia

    INDEX
    Explanations

    subtitles and translations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     volatiles
    0.49
    积
    0.48
     volatile
    0.47
    轆
    0.46
     volatil
    0.45
    megapixel
    0.45
     cadr
    0.45
     посред
    0.45
     Фото
    0.45
     Rope
    0.45
    POSITIVE LOGITS
     subtitles
    0.80
     captions
    0.79
    字幕
    0.76
     Englisch
    0.74
     timing
    0.73
     caption
    0.73
     translations
    0.73
     English
    0.73
    English
    0.71
     subtitle
    0.70
    Act Density 0.012%

    No Known Activations