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

    Neuronpedia

    INDEX
    Explanations

    lists and numbers

    np_max-act · gemini-2.0-flash

    The neuron activates on numbered list markers and outline-style numerals (e.g., “1.”, “2.”, “3.”, etc.) indicating structured list items.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Replica
    -0.07
     salaries
    -0.07
     لها
    -0.07
     채
    -0.07
     GAP
    -0.06
     bids
    -0.06
     yuan
    -0.06
    .conv
    -0.06
    _WIDGET
    -0.06
     Н
    -0.06
    POSITIVE LOGITS
     enlightenment
    0.07
     rég
    0.07
    ucken
    0.07
     activist
    0.07
     clk
    0.06
    .major
    0.06
     Sur
    0.06
    -multi
    0.06
     competence
    0.06
    ductory
    0.06
    Act Density 0.052%

    No Known Activations