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

    Neuronpedia

    INDEX
    Explanations

    November and October dates

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

    tokens representing numbers and dates (numerical values like years, months, times, counts, and other numeric tokens).

    oai_token-act-pair · gpt-5-miniTriggered by @yooniel31
    New Auto-Interp
    Negative Logits
     Summer
    0.82
     July
    0.80
     गर्मियों
    0.80
     June
    0.80
     summer
    0.80
     JULY
    0.79
     SUMMER
    0.79
    JUNE
    0.78
    Summer
    0.77
    June
    0.76
    POSITIVE LOGITS
     November
    1.52
    November
    1.49
    november
    1.47
     november
    1.43
    October
    1.38
    NOVEMBER
    1.36
     October
    1.33
     NOVEMBER
    1.33
     ноябре
    1.32
     novembre
    1.30
    Act Density 0.087%

    No Known Activations