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

    Neuronpedia

    INDEX
    Explanations

    Pro- words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
      
    -3.25
    ↵↵
    -3.19
     
    -2.83
            
    -2.42
     greateſt
    -2.41
    …
    -2.38
          
    -2.34
    人也
    -2.25
    at
    -2.17
    等到
    -2.17
    POSITIVE LOGITS
    ;
    2.78
    彣
    2.77
    اً
    2.72
    桖
    2.67
     Would
    2.67
    erintah
    2.58
    壟
    2.58
     objectif
    2.53
    するので
    2.48
    綃
    2.47
    Act Density 0.019%

    No Known Activations