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

    Neuronpedia

    INDEX
    Explanations

    introducing a concession

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     două
    0.44
     Wuhan
    0.43
     allerdings
    0.43
     några
    0.41
     câteva
    0.41
     QuickBooks
    0.38
     آ
    0.38
     footsteps
    0.38
     însă
    0.37
     bazı
    0.37
    POSITIVE LOGITS
    Nonetheless
    0.71
     Nonetheless
    0.68
     nonetheless
    0.63
    Nevertheless
    0.63
    それでも
    0.61
    Regardless
    0.59
     nevertheless
    0.58
    Still
    0.57
     তবুও
    0.54
     trotzdem
    0.52
    Act Density 0.232%

    No Known Activations