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

    Neuronpedia

    INDEX
    Explanations

    a few

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Huge
    0.75
     oike
    0.67
    が大きい
    0.65
     огромное
    0.64
     alphabetically
    0.64
     Huge
    0.64
    広い
    0.63
    真的
    0.63
    慝
    0.62
    सम्म
    0.61
    POSITIVE LOGITS
     few
    3.75
    few
    3.12
    Few
    3.01
     Few
    2.98
     couple
    2.60
    couple
    2.21
     paar
    2.21
     vài
    2.15
     pocos
    2.10
     몇
    2.03
    Act Density 0.348%

    No Known Activations