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

    Neuronpedia

    INDEX
    Explanations

    Python file paths

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     generalizes
    0.37
    丵
    0.31
     doré
    0.31
    kében
    0.31
    ➚
    0.31
    溦
    0.29
     relatifs
    0.29
    izzati
    0.28
    scheduler
    0.28
     포인트
    0.28
    POSITIVE LOGITS
    Neo
    0.35
     Neo
    0.32
    Neon
    0.27
    Caro
    0.27
    RAM
    0.26
    &
    0.25
     hey
    0.25
    ...\...\...\...\
    0.25
     Hey
    0.25
    yr
    0.24
    Act Density 0.001%

    No Known Activations