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

    Neuronpedia

    INDEX
    Explanations

    we just

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     számos
    1.09
     אשר
    0.85
     zahlreiche
    0.84
     erforder
    0.82
    ใด
    0.81
     המת
    0.79
     označ
    0.79
     ensimmä
    0.78
     sämt
    0.78
     كافة
    0.78
    POSITIVE LOGITS
     keep
    1.16
     get
    1.16
     be
    1.07
     figure
    1.00
     feel
    0.94
     just
    0.89
     really
    0.86
     go
    0.86
     make
    0.85
     KEEP
    0.85
    Act Density 0.786%

    No Known Activations