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

    Neuronpedia

    INDEX
    Explanations

    language and specific words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    frc
    0.81
     പദ്ധതി
    0.79
    '_{
    0.78
     @
    0.76
     spolu
    0.76
     "@
    0.75
    冷的
    0.74
    งาน
    0.73
     embarrassing
    0.72
    ضور
    0.72
    POSITIVE LOGITS
    ――
    0.79
    Japanese
    0.77
     language
    0.73
     kleiner
    0.71
    ──
    0.71
    ---
    0.71
     Japanese
    0.70
    さまざ
    0.70
    language
    0.69
     japonesa
    0.69
    Act Density 0.003%

    No Known Activations