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

    Neuronpedia

    INDEX
    Explanations

    at

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on numeric tokens and numeric data (years, counts, measurements) in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    -speaking
    -0.07
     Charm
    -0.06
     Stephanie
    -0.06
    _ptrs
    -0.06
     güney
    -0.06
    巴
    -0.06
    depart
    -0.06
     para
    -0.06
     preventative
    -0.06
     Ba
    -0.06
    POSITIVE LOGITS
     at
    0.07
    `
↵
    0.07
    ufact
    0.06
    ialias
    0.06
    <=$
    0.06
     Fres
    0.06
    ']));
    0.06
    '];
    0.06
    ,readonly
    0.06
    ./
    0.06
    Act Density 0.061%

    No Known Activations