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

    Neuronpedia

    INDEX
    Explanations

    Time and date

    np_max-act · gemini-2.0-flash

    The neuron is picking up numeric time‐duration or timestamp tokens (e.g. “05:03”, “02:44”, “01:21:46”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Uph
    -0.07
     Immun
    -0.06
    Parm
    -0.06
     spray
    -0.06
     Elle
    -0.06
     gives
    -0.06
     revert
    -0.06
     testName
    -0.06
     teamwork
    -0.06
     transf
    -0.06
    POSITIVE LOGITS
    03
    0.07
    .actor
    0.07
    05
    0.07
    方
    0.07
    :
    0.06
     ;;↵
    0.06
    school
    0.06
    .currentState
    0.06
    道
    0.06
     했다
    0.06
    Act Density 0.020%

    No Known Activations