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

    Neuronpedia

    INDEX
    Explanations

    numeric values and mathematical operations

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

    Numbers, especially decimals or fractions

    np_acts-logits-general · gemini-2.0-flash

    numbers and quantities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Thirty
    -0.94
    thirty
    -0.89
     thirty
    -0.88
     Forty
    -0.88
     forty
    -0.87
    Thirty
    -0.87
     sixty
    -0.82
     Twenty
    -0.81
    teenth
    -0.80
    forty
    -0.80
    POSITIVE LOGITS
    Three
    0.80
    three
    0.75
     three
    0.75
     Three
    0.75
     five
    0.72
    Five
    0.69
     four
    0.69
     trois
    0.67
     Five
    0.67
    five
    0.66
    Act Density 0.432%

    No Known Activations