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

    Neuronpedia

    INDEX
    Explanations

    informational purposes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     highlighting
    0.86
    การ
    0.80
     offending
    0.79
     annoying
    0.76
     disorder
    0.75
     edges
    0.74
     highlight
    0.73
     e
    0.73
     invers
    0.72
     exposed
    0.72
    POSITIVE LOGITS
    Polynomial
    1.06
    tive
    0.98
    Authentication
    0.98
     Businessman
    0.97
    altezza
    0.96
    anía
    0.96
    Authent
    0.95
    Technische
    0.95
    achtige
    0.92
     والمع
    0.92
    Act Density 0.012%

    No Known Activations