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

    Neuronpedia

    INDEX
    Explanations

    when I perceive logic

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    dgs
    0.41
    Interior
    0.40
    activated
    0.37
    afat
    0.37
    ensing
    0.36
    APPROVED
    0.36
    নিষ
    0.36
    )--
    0.36
    rapes
    0.35
    rat
    0.35
    POSITIVE LOGITS
     телефо
    0.44
    стояние
    0.40
     exemple
    0.40
     delivered
    0.39
    तम
    0.38
     기대
    0.38
     nh
    0.37
     Vermeer
    0.36
    Code
    0.36
     entrega
    0.36
    Act Density 0.067%

    No Known Activations