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

    Neuronpedia

    INDEX
    Explanations

    chatbot system prompts

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     ratio
    -0.09
     Ratio
    -0.09
     potency
    -0.09
     curly
    -0.08
    _ratio
    -0.08
     공격
    -0.08
    ceau
    -0.08
     제거
    -0.08
     공급
    -0.08
     लगाने
    -0.08
    POSITIVE LOGITS
     rotina
    0.13
     cotidiano
    0.13
     quotidienne
    0.12
     bustling
    0.12
    午
    0.12
     mornings
    0.12
    每天
    0.12
     बित
    0.12
     morning
    0.12
    忙
    0.12
    Act Density 0.054%

    No Known Activations