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

    Neuronpedia

    INDEX
    Explanations

    section headers and lists

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     renversement
    0.48
    ﺶ
    0.43
     splic
    0.42
     tomates
    0.42
    MenuActive
    0.41
     erasure
    0.41
     bisog
    0.40
    定理
    0.40
     policeman
    0.40
     teorema
    0.40
    POSITIVE LOGITS
    ↵↵
    0.77
    ↵
    0.72
      
    0.66
    ↵↵↵
    0.66
     These
    0.58
     Therefore
    0.54
       
    0.54
    an
    0.51
     This
    0.50
     Thus
    0.50
    Act Density 0.001%

    No Known Activations