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

    Neuronpedia

    INDEX
    Explanations

    bond

    np_max-act · gemini-2.0-flash

    The neuron responds to emotionally charged descriptive words—especially adjectives and adverbs that qualify the depth or intensity of feelings.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     pill
    -0.07
     ilaç
    -0.07
     fluffy
    -0.07
     pyt
    -0.07
    ωσε
    -0.06
     meals
    -0.06
     initializes
    -0.06
     Evening
    -0.06
     orb
    -0.06
     pills
    -0.06
    POSITIVE LOGITS
    ，可
    0.07
     Connection
    0.06
     notamment
    0.06
     Subjects
    0.06
    .
    0.06
     Mem
    0.06
    เอ
    0.06
    intree
    0.06
     moss
    0.06
     addict
    0.06
    Act Density 0.025%

    No Known Activations