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

    Neuronpedia

    INDEX
    Explanations

    answer, verify, decrypt, discover

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    的這個
    0.31
    与其
    0.30
    ondissement
    0.30
    般的
    0.28
    的就是
    0.28
     جدا
    0.28
    odo
    0.27
     Dado
    0.26
     ज्यादातर
    0.26
     عليكم
    0.26
    POSITIVE LOGITS
     them
    0.45
    them
    0.38
     тях
    0.35
     देम
    0.34
    这一切
    0.33
     quietly
    0.32
    everything
    0.31
    🙂
    0.30
     spectacularly
    0.30
     things
    0.30
    Act Density 0.253%

    No Known Activations