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

    Neuronpedia

    INDEX
    Explanations

    conversation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    к
    0.50
    ло
    0.48
    もの
    0.47
    ро
    0.47
    мо
    0.47
    ну
    0.46
    は何
    0.45
    ту
    0.44
    soever
    0.44
    ة
    0.43
    POSITIVE LOGITS
     अबाउट
    0.58
    s
    0.57
     عنها
    0.51
     گفتگو
    0.51
    \}\
    0.50
     chats
    0.49
     conversación
    0.49
     conversations
    0.49
    ों
    0.48
    sock
    0.48
    Act Density 0.716%

    No Known Activations