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

    Neuronpedia

    INDEX
    Explanations

    names in parentheses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Cambrian
    0.67
     balanced
    0.66
     fleece
    0.66
     bouquet
    0.63
     objectively
    0.63
     metamorphic
    0.62
     cobbl
    0.62
     objective
    0.62
     SAG
    0.62
     Sag
    0.61
    POSITIVE LOGITS
     DD
    1.00
    bD
    0.91
     DMD
    0.90
     DZ
    0.90
    Dt
    0.86
    DD
    0.86
    DK
    0.85
     Ди
    0.85
     DP
    0.84
    DDD
    0.83
    Act Density 0.281%

    No Known Activations