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

    Neuronpedia

    INDEX
    Explanations

    unnatural language

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     maiden
    -0.10
     fearing
    -0.08
    (native
    -0.08
     Liter
    -0.08
    �
    -0.08
     autot
    -0.08
    (matrix
    -0.08
    泥
    -0.08
    'class
    -0.08
    =start
    -0.07
    POSITIVE LOGITS
     fictional
    0.08
     fər
    0.08
    .constraint
    0.07
    0.07
    .characters
    0.07
     prank
    0.07
    osomal
    0.07
     faoin
    0.07
    Du
    0.07
     جاتا
    0.07
    Act Density 0.001%
    No Known Activations