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

    Neuronpedia

    INDEX
    Explanations

    removal/relief

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     WR
    -0.08
     caution
    -0.08
     paw
    -0.07
    ौर
    -0.07
     THEM
    -0.07
     스타일
    -0.07
     rö
    -0.07
    .WR
    -0.07
     Piet
    -0.07
     slight
    -0.07
    POSITIVE LOGITS
     eliminates
    0.12
     khỏi
    0.11
     elimina
    0.11
     избав
    0.11
     eliminar
    0.11
     eliminating
    0.11
     erad
    0.11
     eliminate
    0.11
     elimin
    0.11
     দূ
    0.10
    Act Density 0.165%

    No Known Activations