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

    Neuronpedia

    INDEX
    Explanations

    substituting values into formulas

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Carthage
    0.38
    ग्गा
    0.38
     invoices
    0.36
    Clar
    0.36
    èbre
    0.36
     savages
    0.35
     scarves
    0.35
    ifton
    0.35
     расти
    0.35
     ngươi
    0.35
    POSITIVE LOGITS
    -(-
    0.45
    KU
    0.38
    (-\
    0.36
     معنا
    0.35
    (-
    0.34
    යා
    0.34
    mmol
    0.34
    BEN
    0.34
    usu
    0.33
     substitution
    0.33
    Act Density 0.024%

    No Known Activations