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

    Neuronpedia

    INDEX
    Explanations

    [

    np_max-act · gemini-2.0-flash

    This neuron detects occurrences of the opening square‐bracket token “[” (e.g. in citations, markdown links, or section refs).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    -id
    -0.06
    立
    -0.06
     двад
    -0.06
     emacs
    -0.06
    lies
    -0.06
     If
    -0.06
     showdown
    -0.06
    iga
    -0.06
     bailout
    -0.06
    ied
    -0.06
    POSITIVE LOGITS
     зі
    0.07
     perc
    0.06
    :S
    0.06
     угл
    0.06
    .Annotations
    0.06
    []){↵
    0.06
    lodash
    0.06
    kits
    0.06
    _hash
    0.06
     freel
    0.06
    Act Density 0.017%

    No Known Activations