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

    Neuronpedia

    INDEX
    Explanations

    equilibrium

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     маш
    -0.08
    (constants
    -0.07
     Armenian
    -0.07
    蹄
    -0.07
     consort
    -0.07
    泖
    -0.07
     lashes
    -0.07
    缩水
    -0.07
     pagar
    -0.07
     tear
    -0.07
    POSITIVE LOGITS
    ifdef
    0.08
    lat
    0.07
    也没
    0.07
     effects
    0.06
    会让
    0.06
     meu
    0.06
    앖
    0.06
    indexes
    0.06
    㽏
    0.06
    학교
    0.06
    Act Density 0.003%

    No Known Activations