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

    Neuronpedia

    INDEX
    Explanations

    announcements

    np_max-act · gemini-2.0-flash

    The neuron detects announcement language—especially the word “announce” (and excited “we’re excited to announce”‐style phrases) indicating upcoming events or news.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Strategy
    -0.07
    Board
    -0.07
    led
    -0.07
    LED
    -0.07
    .tcp
    -0.06
     Fish
    -0.06
     providing
    -0.06
    áhnout
    -0.06
    isEmpty
    -0.06
    Perhaps
    -0.06
    POSITIVE LOGITS
     $_[
    0.07
    agem
    0.06
     inbox
    0.06
    textTheme
    0.06
    吨
    0.06
    _^(
    0.06
     تهیه
    0.06
     dictionaryWith
    0.06
     ＼:
    0.06
     TTC
    0.06
    Act Density 0.062%

    No Known Activations