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

    Neuronpedia

    INDEX
    Explanations

    you'd like to try another

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     daca
    1.08
     if
    1.08
    如果你
    1.05
     если
    1.03
     dacă
    1.01
    if
    1.00
     jika
    1.00
    如果您
    0.98
    If
    0.97
    ामध्ये
    0.93
    POSITIVE LOGITS
     idiots
    0.85
     jedem
    0.84
     offrent
    0.81
     undermines
    0.80
     councillors
    0.80
     거의
    0.80
     unfold
    0.80
     없음
    0.80
     joht
    0.79
     čast
    0.78
    Act Density 0.087%

    No Known Activations