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

    Neuronpedia

    INDEX
    Explanations

    advantages and features

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    euillez
    -0.07
    upon
    -0.07
    Raw
    -0.07
    Χ
    -0.07
     whispers
    -0.07
    -radio
    -0.06
     pb
    -0.06
    ('=
    -0.06
    component
    -0.06
     );
↵
↵
    -0.06
    POSITIVE LOGITS
    OUNCE
    0.07
     galer
    0.06
    discover
    0.06
     OPT
    0.06
    雙
    0.06
    -gen
    0.06
    interesting
    0.06
    νομα
    0.06
    gens
    0.06
     OMX
    0.06
    Act Density 0.092%

    No Known Activations