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

    Neuronpedia

    INDEX
    Explanations

    remain, vote, leave, referendum, British

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ა
    0.71
    z
    0.66
    é
    0.63
    h
    0.61
     multimillion
    0.60
    elte
    0.58
    ie
    0.57
    á
    0.55
    ت
    0.54
     миллиар
    0.52
    POSITIVE LOGITS
    Т
    0.57
     தங்கள்
    0.54
     for
    0.52
     Ist
    0.52
    時の
    0.51
    Кла
    0.51
    ımı
    0.50
    ボ
    0.50
    াদের
    0.50
    টি
    0.49
    Act Density 0.003%

    No Known Activations