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

    Neuronpedia

    INDEX
    Explanations

    процессы и команды

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    彥
    0.61
    ﮈ
    0.60
    兩
    0.59
     iunie
    0.59
    larghezza
    0.59
    ﮐ
    0.58
     對
    0.58
    ڑھ
    0.57
    ɳ
    0.57
    듈
    0.57
    POSITIVE LOGITS
     не
    1.02
     в
    1.00
     по
    0.98
     на
    0.96
     за
    0.95
     при
    0.95
     про
    0.93
     у
    0.90
     до
    0.88
     с
    0.86
    Act Density 0.302%

    No Known Activations