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

    Neuronpedia

    INDEX
    Explanations

    no hop, meaning, 3 questions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    s
    0.64
    یکیشن
    0.48
    ্টর
    0.46
     ridiculed
    0.46
    ség
    0.45
     frequent
    0.43
    rijk
    0.43
     wretched
    0.43
    ের
    0.43
     hateful
    0.42
    POSITIVE LOGITS
    າດ
    0.51
    ቃል
    0.48
    ניה
    0.48
    Producer
    0.44
    ברה
    0.44
     Cuenta
    0.44
    Ist
    0.44
    鈮
    0.44
    У
    0.43
    Honeycomb
    0.42
    Act Density 0.002%

    No Known Activations