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

    Neuronpedia

    INDEX
    Explanations

    safety, respect, and care

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     repur
    0.87
     Retry
    0.83
     repurposed
    0.80
     prefabricated
    0.79
     uptick
    0.78
     incentivize
    0.78
     PTSD
    0.76
     KPIs
    0.76
     sparen
    0.75
     disgruntled
    0.75
    POSITIVE LOGITS
    respect
    0.80
    Equity
    0.79
     affirming
    0.79
     Affirm
    0.75
     Equity
    0.75
    intel
    0.75
     EQUITY
    0.75
     respect
    0.74
    affirming
    0.74
     лично
    0.73
    Act Density 0.237%

    No Known Activations