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

    Neuronpedia

    INDEX
    Explanations

    regex questions

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on digits and numeric quantifier syntax (e.g. numbers inside `{ }`, repetitions, and related numeric tokens) within regular‐expression text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _rd
    -0.07
    Workflow
    -0.06
    Cache
    -0.06
    _RD
    -0.06
    _Abstract
    -0.06
    -dd
    -0.06
    (scope
    -0.06
    odb
    -0.06
    _final
    -0.06
    GLOBALS
    -0.06
    POSITIVE LOGITS
     Xiao
    0.06
     tier
    0.06
     granite
    0.06
     tome
    0.06
    structures
    0.06
     Moments
    0.06
     ENG
    0.06
     vais
    0.06
    CKET
    0.06
     정치
    0.06
    Act Density 0.030%

    No Known Activations