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

    Neuronpedia

    INDEX
    Explanations

    against + noun

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
      
    -1.31
     is
    -1.13
        
    -1.12
     to
    -1.11
     You
    -1.08
     
    -1.06
       
    -1.06
     into
    -1.05
     If
    -1.05
         
    -1.03
    POSITIVE LOGITS
     hunne
    1.14
     bileklik
    1.11
     すぐ
    1.10
    CHREIB
    1.09
     ausein
    1.09
     deseos
    1.06
    已經
    1.03
     zimowa
    1.02
    已经
    1.02
    裢
    1.01
    Act Density 0.038%

    No Known Activations