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

    Neuronpedia

    INDEX
    Explanations

    months followed by day and year

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

    This neuron detects date expressions—especially month names and nearby numeric tokens that form calendar dates.

    oai_token-act-pair · gpt-5-miniTriggered by @yooniel31
    New Auto-Interp
    Negative Logits
     excitatory
    0.32
     implies
    0.31
    லோச
    0.31
     ponder
    0.30
     excret
    0.29
    ణులు
    0.29
     waveguides
    0.28
     CCSDT
    0.28
     meditative
    0.28
     terpen
    0.28
    POSITIVE LOGITS
     February
    0.66
     January
    0.66
     September
    0.64
     April
    0.61
     November
    0.61
    September
    0.61
    February
    0.61
     July
    0.60
     December
    0.59
     June
    0.58
    Act Density 0.015%

    No Known Activations