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

    Neuronpedia

    INDEX
    Explanations

    test scenarios and correctness

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    clonal
    0.44
    GRANT
    0.42
    Appro
    0.39
     приблизи
    0.39
    Approx
    0.38
    physema
    0.38
    оте
    0.38
    лье
    0.38
    прода
    0.38
    τεί
    0.38
    POSITIVE LOGITS
     correctness
    0.60
     tests
    0.59
     symmetries
    0.59
     robustness
    0.58
     trigonometric
    0.58
     symmetry
    0.56
     tested
    0.52
     correctly
    0.52
     cases
    0.51
    错误
    0.51
    Act Density 0.317%

    No Known Activations