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

    Neuronpedia

    INDEX
    Explanations

    tax penalties

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     sağ
    0.46
     descendants
    0.46
     plonge
    0.46
     welded
    0.45
     trast
    0.45
     fuselage
    0.45
     dout
    0.45
     forged
    0.44
     faction
    0.44
     plastica
    0.44
    POSITIVE LOGITS
     expecting
    0.44
     NEAR
    0.42
    EXPECT
    0.41
    innis
    0.41
    兼
    0.40
    Pets
    0.37
    均
    0.37
    Prob
    0.37
    PSR
    0.37
     EXPECT
    0.36
    Act Density 0.000%

    No Known Activations