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

    Neuronpedia

    INDEX
    Explanations

    internal thought and imagination

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     kræ
    0.89
     signatories
    0.88
     intolerance
    0.83
     erhalten
    0.81
     dokument
    0.80
     mandate
    0.78
     adhered
    0.77
     균
    0.77
     melaksanakan
    0.76
     couronne
    0.76
    POSITIVE LOGITS
     fantas
    1.07
     dream
    1.05
     memories
    1.00
     imagining
    1.00
     memory
    0.99
    memory
    0.97
     imagines
    0.96
     imaginary
    0.96
     Imag
    0.93
     remembering
    0.93
    Act Density 0.760%

    No Known Activations