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

    Neuronpedia

    INDEX
    Explanations

    Unstructured text snippets

    np_max-act · gemini-2.0-flash

    The neuron activates on tokens that signal totality or exhaustiveness—words or punctuation that mark something as complete or all-encompassing.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    statements related to legal processes and tax Appeals.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Ordinary
    -0.06
     relatively
    -0.06
    session
    -0.06
    �
    -0.06
    _sub
    -0.06
    イント
    -0.06
    详情
    -0.06
    First
    -0.06
     tub
    -0.06
     recession
    -0.06
    POSITIVE LOGITS
    Correo
    0.07
     перепис
    0.07
    prix
    0.06
    _epi
    0.06
    "%(
    0.06
    contri
    0.06
     dobr
    0.06
     poisoning
    0.06
     hippoc
    0.06
    düğ
    0.06
    Act Density 0.140%

    No Known Activations