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

    Neuronpedia

    INDEX
    Explanations

    character descriptions

    np_max-act · gemini-2.0-flash

    The neuron fires on bits of text that look like character–profile fields or personal‐bio details (e.g. headings like “Occupation,” “Physical description,” “Personality traits,” or the ensuing descriptive values).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    character details and traits in creative writing contexts.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     lẽ
    -0.07
    ΑΣ
    -0.06
     subclasses
    -0.06
    pectrum
    -0.06
    .im
    -0.06
     contrace
    -0.06
     Watches
    -0.06
     indoor
    -0.06
    ezier
    -0.06
    ække
    -0.06
    POSITIVE LOGITS
    чна
    0.07
     nichts
    0.07
     adapted
    0.06
     slip
    0.06
    uating
    0.06
     repaired
    0.06
     Rac
    0.06
     Instructor
    0.06
     konci
    0.06
    ляют
    0.06
    Act Density 0.042%

    No Known Activations