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

    Neuronpedia

    INDEX
    Explanations

    or before

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     zoos
    0.79
     Separ
    0.75
     helps
    0.68
     पार्ट्स
    0.67
    र्द
    0.67
     borders
    0.66
     Entry
    0.66
    गिंग
    0.65
    زء
    0.65
     यार
    0.65
    POSITIVE LOGITS
    before
    1.24
     before
    1.16
    Before
    1.08
    BEFORE
    1.03
     sebelum
    1.00
     decisione
    0.96
     antes
    0.95
     BEFORE
    0.94
     înainte
    0.94
     Before
    0.93
    Act Density 0.002%

    No Known Activations