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

    Neuronpedia

    INDEX
    Explanations

    Danish, Deep, Diamond, DLA, DMI

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     fleece
    0.67
     Cambrian
    0.65
     Sag
    0.64
     cardinal
    0.64
    ったり
    0.64
     metamorphic
    0.63
     Unterkunft
    0.60
    Sag
    0.60
     dign
    0.60
     bouquet
    0.60
    POSITIVE LOGITS
     DD
    1.00
    bD
    0.97
     DMD
    0.93
     DZ
    0.91
     DP
    0.89
     DDS
    0.86
     DX
    0.85
     Ди
    0.84
     DDD
    0.84
    DK
    0.84
    Act Density 0.414%

    No Known Activations