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

    Neuronpedia

    INDEX
    Explanations

    media entities and sources

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     boasting
    0.41
     cujo
    0.40
    给我
    0.38
    oodles
    0.38
    是我
    0.37
    人々
    0.37
    全是
    0.37
     imaju
    0.37
    被称为
    0.37
    在我
    0.36
    POSITIVE LOGITS
     correspondent
    0.68
     correspondents
    0.67
     readers
    0.63
     understands
    0.61
    注意到
    0.61
     witnessed
    0.60
     Correspondent
    0.57
     lecteurs
    0.57
     Editors
    0.56
     recommends
    0.56
    Act Density 0.006%

    No Known Activations