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

    Neuronpedia

    INDEX
    Explanations

    family members

    np_max-act · gemini-2.0-flash

    terms related to familial relationships, particularly focusing on grandparents and their interactions or roles within families.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron activates on mentions of “grandparent” (including “grandparents,” “grandmother,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Acts
    -0.07
     menu
    -0.07
     lig
    -0.07
     Menu
    -0.07
     firms
    -0.06
    Pad
    -0.06
    Ent
    -0.06
     Soul
    -0.06
     fuse
    -0.06
     отдель
    -0.06
    POSITIVE LOGITS
     grandma
    0.12
     grandmother
    0.11
     grandfather
    0.10
     grandparents
    0.09
     Grandma
    0.09
     GM
    0.07
    ые
    0.07
     وغير
    0.07
    átka
    0.07
     #=>
    0.06
    Act Density 0.006%

    No Known Activations