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

    Neuronpedia

    INDEX
    Explanations

    years and numbers

    np_acts-logits-general · gemini-2.5-flash-lite

    This neuron detects year-like numeric tokens (four-digit years / date numbers) in the text.

    oai_token-act-pair · gpt-5-miniTriggered by @yooniel31
    New Auto-Interp
    Negative Logits
    ０
    0.99
    h
    0.93
    と思います
    0.93
    اء
    0.89
    های
    0.88
     Investigative
    0.87
    是很
    0.86
    0.86
     I
    0.85
    ó
    0.84
    POSITIVE LOGITS
    स
    1.27
     silam
    1.13
    ي
    1.12
    ی
    1.09
    에
    1.07
     года
    1.05
    itare
    1.05
    酯
    1.02
     sistemat
    1.00
    อัต
    1.00
    Act Density 0.107%

    No Known Activations