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

    Neuronpedia

    INDEX
    Explanations

    say "cur" word

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    uttgart
    -0.07
     chapter
    -0.07
    sword
    -0.07
    orth
    -0.07
    .sky
    -0.06
    letion
    -0.06
    permit
    -0.06
     Phen
    -0.06
    printed
    -0.06
    .PrimaryKey
    -0.06
    POSITIVE LOGITS
    "user
    0.07
    ewis
    0.07
     disag
    0.06
     tox
    0.06
     trab
    0.06
     Вик
    0.06
     fazla
    0.06
    (fi
    0.06
     أجل
    0.06
    ับ
    0.06
    Act Density 0.048%

    No Known Activations