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

    Neuronpedia

    INDEX
    Explanations

    specific numeric values and scientific identifiers

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Numbers of things (four, 4, tetra)

    np_acts-logits-general · gemini-2.0-flash

    four or five quantities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Ambos
    -0.70
    Kedua
    -0.70
     Ambos
    -0.63
     Beide
    -0.63
     Both
    -0.63
     दोनों
    -0.62
    Both
    -0.62
     both
    -0.59
    both
    -0.57
    BOTH
    -0.56
    POSITIVE LOGITS
     four
    2.20
    four
    1.93
     five
    1.92
    five
    1.70
     Four
    1.70
    Four
    1.69
     quatre
    1.69
     quatro
    1.65
     FOUR
    1.58
     cuatro
    1.55
    Act Density 0.342%

    No Known Activations