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

    Neuronpedia

    INDEX
    Explanations

    HTML/CSS: showing or active elements

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    _unit
    -0.07
    creating
    -0.06
     greet
    -0.06
     unhappy
    -0.06
    (pages
    -0.06
    Pro
    -0.06
    observ
    -0.06
    Unit
    -0.06
     руч
    -0.06
    <lemma
    -0.06
    POSITIVE LOGITS
    phy
    0.07
     Polic
    0.07
     мо
    0.07
    аніт
    0.07
    ....
    0.06
    ilog
    0.06
     serialize
    0.06
     बढ़
    0.06
     shorts
    0.06
    _MIDDLE
    0.06
    Act Density 0.008%

    No Known Activations