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

    Neuronpedia

    INDEX
    Explanations

    phase 1 or chapter 1

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     secondly
    0.46
     second
    0.45
    second
    0.45
    considering
    0.41
     segunda
    0.41
     deuxième
    0.40
     druhé
    0.39
     tweede
    0.39
    ivé
    0.38
     hierarchical
    0.38
    POSITIVE LOGITS
     Breakdown
    0.66
     chapters
    0.59
     Chapter
    0.56
     Kapitel
    0.55
     breakdown
    0.55
     chapter
    0.54
     CHAPTER
    0.54
    Chapter
    0.53
     Chapters
    0.48
    概述
    0.48
    Act Density 0.041%

    No Known Activations