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

    Neuronpedia

    INDEX
    Explanations

    with, without, avec, with

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     deoarece
    0.38
    hmC
    0.35
     aquilo
    0.34
     joten
    0.34
    hatan
    0.34
    itates
    0.33
     lop
    0.33
     pois
    0.33
    hluk
    0.33
     deshalb
    0.32
    POSITIVE LOGITS
     причем
    0.74
     עם
    0.68
     avec
    0.63
     WITH
    0.61
     with
    0.60
    ただし
    0.57
     включая
    0.57
     without
    0.57
    하거나
    0.57
     χωρίς
    0.55
    Act Density 1.786%

    No Known Activations