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

    Neuronpedia

    INDEX
    Explanations

    summarizes things

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     included
    0.47
     include
    0.43
    included
    0.43
     típ
    0.42
     Included
    0.42
    Included
    0.41
    ariales
    0.40
     포함
    0.40
     imprinted
    0.39
     Include
    0.39
    POSITIVE LOGITS
     summarizes
    0.46
     summary
    0.45
     SUMMARY
    0.44
     کتاب
    0.43
    Summary
    0.42
    SUMMARY
    0.42
     summar
    0.42
     Almost
    0.42
    summar
    0.41
     books
    0.41
    Act Density 0.001%

    No Known Activations