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

    Neuronpedia

    INDEX
    Explanations

    Then

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     dept
    0.42
    !!!!!!!!
    0.42
     hesitation
    0.42
     hypoth
    0.42
    !!!!!!!
    0.42
     lepton
    0.42
    มีความ
    0.41
     applic
    0.41
    тил
    0.41
     möchte
    0.41
    POSITIVE LOGITS
     And
    0.58
     Lately
    0.58
     Worse
    0.54
    And
    0.51
    But
    0.51
     Именно
    0.51
     Today
    0.50
     население
    0.50
     Сьогодні
    0.49
    但这
    0.49
    Act Density 0.007%

    No Known Activations