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

    Neuronpedia

    INDEX
    Explanations

    waiting

    np_max-act · gemini-2.0-flash

    narratives that involve personal experiences and expressions of emotional states related to self-reflection and vulnerability.

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

    The neuron is picking up on the little “filler” and self-referential bits of first-person narration—words like “I,” “just,” “’ll,” “bear,” “embarrassment,” etc., that mark personal, introspective narrative.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     sanctioned
    -0.08
    brains
    -0.07
    .mutex
    -0.07
     prized
    -0.06
     Broad
    -0.06
    Sql
    -0.06
    .card
    -0.06
    _debug
    -0.06
    _pipe
    -0.06
    ить
    -0.06
    POSITIVE LOGITS
    фф
    0.07
     PTS
    0.07
     KW
    0.07
    τι
    0.06
    oppable
    0.06
    \Redirect
    0.06
     Palette
    0.06
     triang
    0.06
    xAA
    0.06
    _countries
    0.06
    Act Density 0.035%

    No Known Activations