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

    Neuronpedia

    INDEX
    Explanations

    thinking step by step

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     TOTAL
    0.46
     TOT
    0.41
     SWT
    0.40
     Total
    0.38
     মোট
    0.38
     STM
    0.38
     अराउंड
    0.36
    TOTAL
    0.35
     TDs
    0.35
     tenders
    0.35
    POSITIVE LOGITS
     thing
    0.51
     coisa
    0.41
    ości
    0.39
     image
    0.39
     cosa
    0.39
     result
    0.38
    thing
    0.37
    Thing
    0.37
     содержимое
    0.37
     dataset
    0.36
    Act Density 0.002%

    No Known Activations