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

    Neuronpedia

    INDEX
    Explanations

    horse racing

    np_max-act · gemini-2.0-flash

    references to specific harness racing events and related figures.

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

    The neuron selectively activates on plural nouns ending in “-ers,” especially those used as names or titles (e.g. Trotters, Tigers, Warriors, Rhinos).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .unsplash
    -0.07
     contains
    -0.07
    uf
    -0.06
    William
    -0.06
     Strip
    -0.06
     Biz
    -0.06
    lesi
    -0.06
    ,status
    -0.06
    Wei
    -0.06
     consist
    -0.06
    POSITIVE LOGITS
     carbohydrate
    0.06
    916
    0.06
    реш
    0.06
    modity
    0.06
    /en
    0.06
    irement
    0.06
    451
    0.06
    reur
    0.06
    ↵      ↵
    0.06
     artifact
    0.06
    Act Density 0.012%

    No Known Activations