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

    Neuronpedia

    INDEX
    Explanations

    active

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    Bank
    -0.07
     haut
    -0.07
    ret
    -0.06
    驚
    -0.06
    voř
    -0.06
     Staten
    -0.06
     Chunk
    -0.06
     alignments
    -0.06
     Jam
    -0.06
    -m
    -0.06
    POSITIVE LOGITS
     superheroes
    0.07
    iltr
    0.06
    ді
    0.06
    (Student
    0.06
    assignment
    0.06
    sess
    0.06
    álně
    0.06
    �니다
    0.06
     Microwave
    0.06
     vibrant
    0.06
    Act Density 0.005%

    No Known Activations