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

    Neuronpedia

    INDEX
    Explanations

    109

    np_max-act · gemini-2.0-flash

    This neuron fires on short subword fragments of chemical compound names—especially pieces of drug names (e.g. ‘cl’, ‘az’, ‘epam’ from “clonazepam,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     washer
    -0.08
    VOICE
    -0.07
    ンデ
    -0.06
    िध
    -0.06
    τερ
    -0.06
    belief
    -0.06
     ofrece
    -0.06
     έργ
    -0.06
     pix
    -0.06
    -res
    -0.06
    POSITIVE LOGITS
    quisites
    0.07
     Lyn
    0.07
    Really
    0.06
    .LayoutStyle
    0.06
     συμπ
    0.06
    :mysql
    0.06
     united
    0.06
     Mickey
    0.06
    uuml
    0.06
     %[
    0.06
    Act Density 0.097%

    No Known Activations