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

    Neuronpedia

    INDEX
    Explanations

    section headers and list items

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ;
    0.19
     deletes
    0.17
     또는
    0.14
    ↵
    0.14
     they
    0.14
    ,”
    0.14
    ])
    0.13
    any
    0.13
    。”
    0.13
    (),
    0.13
    POSITIVE LOGITS
     Basics
    0.18
    什么是
    0.18
     What
    0.17
     Types
    0.17
    What
    0.17
    Caractéristiques
    0.17
     Roles
    0.17
     Importance
    0.17
     Advantages
    0.17
     기본적인
    0.17
    Act Density 4.703%

    No Known Activations