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

    Neuronpedia

    INDEX
    Explanations

    Parentheses and brackets

    np_max-act · gemini-2.0-flash

    narrative elements related to fantasy and magical themes.

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

    This neuron activates on uppercase title or header tokens and associated metadata (e.g. “BOOK,” “BOOK #3,” “first edition,” “CROWNS OF SILVER & ASH”), effectively spotting section headings or book‐listing headers.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    st
    -0.07
     день
    -0.07
    Light
    -0.06
    :update
    -0.06
    Water
    -0.06
    ST
    -0.06
    부터
    -0.06
     animations
    -0.06
    Def
    -0.06
    ant
    -0.06
    POSITIVE LOGITS
    ịnh
    0.08
     relate
    0.07
     allowable
    0.07
     inexp
    0.06
     Alvarez
    0.06
    ระเบ
    0.06
    mek
    0.06
    itmap
    0.06
    vere
    0.06
     encore
    0.06
    Act Density 0.016%

    No Known Activations