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

    Neuronpedia

    INDEX
    Explanations

    Skill-challenge balance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    "
    0.38
    olen
    0.38
    hetical
    0.38
     Something
    0.38
    াইড
    0.38
    ev
    0.37
    ترم
    0.37
    ethical
    0.37
    tik
    0.37
     Compatible
    0.36
    POSITIVE LOGITS
     melhorar
    0.47
     отзывы
    0.47
     動物
    0.47
     х
    0.47
     Лит
    0.46
     г
    0.46
     গে
    0.46
     улуч
    0.45
    椃
    0.45
     بم
    0.45
    Act Density 0.000%

    No Known Activations