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

    Neuronpedia

    INDEX
    Explanations

    how long, area

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     for
    0.52
     marking
    0.45
     SK
    0.44
    WK
    0.44
     pled
    0.42
    Schwarz
    0.42
     IDC
    0.42
     Ansatz
    0.42
     කර
    0.42
     favorably
    0.42
    POSITIVE LOGITS
     indicadores
    0.44
    覺
    0.43
    険
    0.42
    ুক্তি
    0.42
     mediators
    0.42
     世界
    0.41
    ährung
    0.41
    ്ര
    0.40
     காலம்
    0.40
     则
    0.40
    Act Density 0.001%

    No Known Activations