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

    Neuronpedia

    INDEX
    Explanations

    code and punctuation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ’
    0.52
     et
    0.50
     bohat
    0.42
    î
    0.41
    !
    0.41
     وإ
    0.39
      
    0.38
    '
    0.38
     Gough
    0.38
     But
    0.38
    POSITIVE LOGITS
    :</
    0.49
    的其他
    0.45
    েক্টর
    0.44
    ５
    0.44
    设置为
    0.44
     inoltre
    0.43
     spaceleft
    0.42
    এছ
    0.42
    HideFlags
    0.42
    5
    0.42
    Act Density 0.131%

    No Known Activations