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

    Neuronpedia

    INDEX
    Explanations

    sports results

    np_max-act · gemini-2.0-flash

    The neuron flags tokens in structured sports‐results lines—especially match outcome markers (“def,” “walk-over”) and the accompanying numeric scores.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     hour
    -0.07
     junior
    -0.07
     Wikimedia
    -0.07
    Israel
    -0.06
    unga
    -0.06
    후기
    -0.06
    atomy
    -0.06
     Year
    -0.06
    arians
    -0.06
    956
    -0.06
    POSITIVE LOGITS
    .Maximum
    0.07
     khách
    0.07
    abbo
    0.07
    権
    0.07
    सम
    0.07
     ",↵
    0.07
    -New
    0.06
    Processing
    0.06
    �
    0.06
    матрива
    0.06
    Act Density 0.009%

    No Known Activations