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

    Neuronpedia

    INDEX
    Explanations

    Intense scenarios

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on words describing hugs or embraces—i.e. tokens denoting physical, affectionate embraces.

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

    scenes depicting emotional connections and intimate relationships between characters.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    oxetine
    -0.07
     Конститу
    -0.07
     روی
    -0.06
    -map
    -0.06
     سور
    -0.06
    basename
    -0.06
    workflow
    -0.06
    experiment
    -0.06
    ��이
    -0.06
    plings
    -0.06
    POSITIVE LOGITS
     CPPUNIT
    0.07
     Tina
    0.06
     Twins
    0.06
    ederland
    0.06
    ..<
    0.06
    Malloc
    0.06
     Drain
    0.06
     maken
    0.06
    ngle
    0.06
    Filed
    0.06
    Act Density 0.080%

    No Known Activations