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

    Neuronpedia

    INDEX
    Explanations

    -

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on numeric tokens—especially floating‐point numbers and signed decimals—in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Sep
    -0.08
     Describe
    -0.07
     acesso
    -0.07
    PFN
    -0.07
     друга
    -0.07
    ียรต
    -0.07
    Court
    -0.06
    ARTH
    -0.06
     USC
    -0.06
     BUILD
    -0.06
    POSITIVE LOGITS
     -
    0.09
    )—
    0.07
    "-
    0.07
     않았다
    0.07
    -
    0.07
    '-
    0.07
    员
    0.07
    -п
    0.07
     indemn
    0.06
    ‑
    0.06
    Act Density 0.008%

    No Known Activations