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

    Neuronpedia

    INDEX
    Explanations

    descriptive adjectives or specific terms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Y
    1.18
     M
    1.15
     Z
    1.13
     P
    1.08
     E
    1.07
     F
    1.07
     T
    1.07
     R
    1.07
     K
    1.06
    Z
    1.04
    POSITIVE LOGITS
     socalled
    1.27
     famously
    1.27
     cosidd
    1.27
     sogen
    1.19
     tzv
    1.19
    quefois
    1.15
     """"
    1.15
    roversial
    1.15
     dakkh
    1.14
     berühm
    1.14
    Act Density 2.636%

    No Known Activations