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

    Neuronpedia

    INDEX
    Explanations

    spatial arrangements and precise placement

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    })^{
    0.73
     exacerbated
    0.72
     debounce
    0.66
     inéd
    0.65
     фактически
    0.64
     amelior
    0.63
     coalesce
    0.63
     década
    0.62
     nghĩa
    0.62
     clogging
    0.61
    POSITIVE LOGITS
     evenly
    0.98
     carefully
    0.96
     parallel
    0.94
     Parallel
    0.91
    Parallel
    0.90
    parallel
    0.87
     overlapping
    0.87
     about
    0.85
     Carefully
    0.85
    about
    0.85
    Act Density 0.102%

    No Known Activations