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

    Neuronpedia

    INDEX
    Explanations

    pretty

    np_max-act · gemini-2.0-flash

    references to personal experiences and opinions regarding beauty products.

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

    This neuron responds to the use of the adverb “pretty” as an intensifier before an adjective.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Huawei
    -0.07
     furthermore
    -0.07
    562
    -0.07
    112
    -0.07
     Force
    -0.07
     Allow
    -0.07
    028
    -0.07
    ASE
    -0.07
     Alien
    -0.07
    _SL
    -0.06
    POSITIVE LOGITS
     pretty
    0.15
     Pretty
    0.14
    Pretty
    0.13
    pretty
    0.12
     petty
    0.09
     fair
    0.08
    PRETTY
    0.08
    ppy
    0.07
    피
    0.07
     předpis
    0.07
    Act Density 0.008%

    No Known Activations