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

    Neuronpedia

    INDEX
    Explanations

    internet forum/code snippets

    np_max-act · gemini-2.0-flash

    The neuron primarily fires on the special end-of-text token (“<|eot_id|>”), i.e. it detects segment-boundary markers.

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

    technical terminology related to programming and data structures.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     ​
    -0.07
     insecure
    -0.07
    /head
    -0.06
    bsd
    -0.06
    .up
    -0.06
    aging
    -0.06
    泉
    -0.06
    ла
    -0.06
     submitted
    -0.06
     missing
    -0.06
    POSITIVE LOGITS
    дом
    0.07
     altercation
    0.06
     Clerk
    0.06
    游
    0.06
    _answer
    0.06
     �
    0.06
    介
    0.06
     �
    0.06
    ีพ
    0.06
    .Hash
    0.06
    Act Density 0.128%

    No Known Activations