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

    Neuronpedia

    INDEX
    Explanations

    that

    np_max-act · gemini-2.0-flash

    references to specific individuals or titles.

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

    The neuron detects indefinite noun‐phrase introductions used to characterize a person or group (i.e. “a [entity] …” with descriptive modifiers).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     softly
    -0.07
    ara
    -0.07
    -flight
    -0.06
    ez
    -0.06
     successfully
    -0.06
    (Level
    -0.06
     lace
    -0.06
    rances
    -0.06
     riders
    -0.06
    embr
    -0.06
    POSITIVE LOGITS
     Nimbus
    0.07
    $__
    0.06
     Assertion
    0.06
     注
    0.06
    产业
    0.06
     pohled
    0.06
     річ
    0.06
    	↵↵
    0.06
    ']*
    0.06
    Stat
    0.06
    Act Density 0.074%

    No Known Activations