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

    Neuronpedia

    INDEX
    Explanations

    proper

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     sammen
    -0.07
    agnost
    -0.07
     presenta
    -0.06
    $fdata
    -0.06
    ==>
    -0.06
    дин
    -0.06
     llama
    -0.06
     contrast
    -0.06
    atively
    -0.06
    düğü
    -0.06
    POSITIVE LOGITS
     properly
    0.09
     proper
    0.09
     Proper
    0.09
    .per
    0.08
    par
    0.07
     prepared
    0.07
     Pure
    0.07
     Superior
    0.07
     Pieces
    0.07
     Parenthood
    0.07
    Act Density 0.018%

    No Known Activations