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

    Neuronpedia

    INDEX
    Explanations

    first item or chapter

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     environments
    0.42
     utilities
    0.40
    దా
    0.39
     indeed
    0.38
     информа
    0.38
    тана
    0.38
     adhesives
    0.37
     internal
    0.37
     statistical
    0.36
     varied
    0.36
    POSITIVE LOGITS
     week
    0.76
     chapter
    0.75
    week
    0.71
     month
    0.66
    month
    0.66
     paragraph
    0.65
     चैप्टर
    0.64
    Week
    0.63
     subparagraph
    0.61
     chapitre
    0.60
    Act Density 0.014%

    No Known Activations