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

    Neuronpedia

    INDEX
    Explanations

    instructions/requests

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     overdue
    -0.07
    ywać
    -0.07
    etc
    -0.07
    לס
    -0.07
    丧
    -0.07
    Ɛ
    -0.06
     '('
    -0.06
    喜剧
    -0.06
     subgroup
    -0.06
    Ǽ
    -0.06
    POSITIVE LOGITS
    _legacy
    0.07
    牡
    0.07
    _signals
    0.07
    _dis
    0.07
    石油
    0.07
    Live
    0.07
    油气
    0.07
     giant
    0.07
     которых
    0.07
     dolphins
    0.07
    Act Density 0.002%

    No Known Activations