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

    Neuronpedia

    INDEX
    Explanations

    raw materials, primary classifications

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ある
    1.13
    ли
    0.96
    大
    0.95
    お
    0.94
    ول
    0.89
    見
    0.89
    なく
    0.84
    多い
    0.82
     is
    0.82
    取り
    0.80
    POSITIVE LOGITS
    ]
    1.16
    )
    1.10
    us
    1.05
    k
    1.05
    ı
    0.98
    ς
    0.96
    ओं
    0.95
    ی
    0.89
     primary
    0.86
     первич
    0.86
    Act Density 0.050%

    No Known Activations