INDEX
    Explanations

    Alpaca Stanford instruction

    New Auto-Interp
    Negative Logits
    0.47
     attracting
    0.43
    мпаваць
    0.41
     되겠죠
    0.41
    Ό
    0.41
     destined
    0.40
     maintaining
    0.40
    ouflage
    0.40
    clusion
    0.39
    你会
    0.38
    POSITIVE LOGITS
     leb
    0.40
     lemongrass
    0.39
     teilweise
    0.38
     상세
    0.38
     lej
    0.37
     monkey
    0.37
     Vergangenheit
    0.37
     szczeg
    0.36
     Shankar
    0.35
     kur
    0.35
    Act Density 0.001%

    No Known Activations