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

    Neuronpedia

    INDEX
    Explanations

    replacement or substitution

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    overl
    0.42
    icates
    0.40
     BufferedWriter
    0.39
    secondary
    0.39
    extras
    0.39
    叠加
    0.38
    ematic
    0.37
    ánh
    0.37
    ẳ
    0.36
    eg
    0.35
    POSITIVE LOGITS
     instead
    1.20
     вместо
    1.04
    代わりに
    1.04
    instead
    1.02
     Instead
    1.00
    Instead
    0.97
     invece
    0.92
     대신
    0.90
     replace
    0.87
     statt
    0.87
    Act Density 0.433%

    No Known Activations