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

    Neuronpedia

    INDEX
    Explanations

    0 or (

    np_max-act · gemini-2.0-flash

    The neuron primarily detects numeric literals (e.g. version numbers, port numbers, numeric arguments) in code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     jež
    -0.07
     Buying
    -0.06
     pedig
    -0.06
     deepest
    -0.06
     distorted
    -0.06
     trận
    -0.06
     pantry
    -0.06
    782
    -0.06
     gx
    -0.06
     renown
    -0.06
    POSITIVE LOGITS
     ''),↵
    0.06
     Pak
    0.06
    (check
    0.06
     $$
    0.06
     tonumber
    0.06
    .gsub
    0.06
    .zh
    0.06
    ilih
    0.06
     <-
    0.06
    原因
    0.06
    Act Density 0.003%

    No Known Activations