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

    Neuronpedia

    INDEX
    Explanations

    dialogue

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on informal greeting questions—particularly “Hey,” at the start of a conversational turn.

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

    informal two-person dialogue lines, especially conversational openings and questions in casual speech.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
    Util
    -0.07
    oder
    -0.07
    28
    -0.07
     disruptions
    -0.06
     INNER
    -0.06
     reform
    -0.06
    groupId
    -0.06
    _MAP
    -0.06
     Inner
    -0.06
     those
    -0.06
    POSITIVE LOGITS
     restrictive
    0.06
    -meter
    0.06
    cí
    0.06
     estates
    0.06
     jmp
    0.06
    _CLASSES
    0.06
    evice
    0.06
    _cpp
    0.06
    (mp
    0.06
    _refs
    0.06
    Act Density 0.098%

    No Known Activations