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

    Neuronpedia

    INDEX
    Explanations

    can introduce

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     سازی
    0.53
     देकर
    0.51
    тоў
    0.50
    ποιη
    0.48
    ल्डर
    0.47
    ską
    0.46
    바일
    0.46
    곸
    0.46
    ਤਾ
    0.46
    ნის
    0.46
    POSITIVE LOGITS
    n
    0.57
     Unfortunately
    0.56
     This
    0.52
     Marriage
    0.49
     Short
    0.49
     To
    0.49
    eres
    0.48
     Template
    0.47
     Because
    0.46
    ne
    0.46
    Act Density 0.004%

    No Known Activations