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

    Neuronpedia

    INDEX
    Explanations

    punctuation marks

    np_max-act · gemini-2.0-flash

    This neuron detects speaker/role labels and dialogue-turn markers (tokens identifying who is speaking or header IDs).

    oai_token-act-pair · gpt-5-miniTriggered by @yooniel31
    New Auto-Interp
    Negative Logits
     Tess
    -0.07
    -services
    -0.07
    (Html
    -0.06
    tras
    -0.06
     Entries
    -0.06
    816
    -0.06
     Pages
    -0.06
    Dimension
    -0.06
     neredeyse
    -0.06
    Documents
    -0.06
    POSITIVE LOGITS
    OLUTION
    0.07
    (username
    0.07
     siyas
    0.06
     جام
    0.06
    @FindBy
    0.06
     Blur
    0.06
     histo
    0.06
    .maximum
    0.06
    acock
    0.06
     ústav
    0.06
    Act Density 0.053%

    No Known Activations