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

    Neuronpedia

    INDEX
    Explanations

    Hidden objects/locations

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     staining
    -0.08
    ийн
    -0.08
    ેપ
    -0.08
     greatness
    -0.08
     aromatic
    -0.07
     trekt
    -0.07
    ,response
    -0.07
     więc
    -0.07
     kæ
    -0.07
     vraagt
    -0.07
    POSITIVE LOGITS
     disguised
    0.15
     guise
    0.15
     disguis
    0.13
     innoc
    0.13
     disguise
    0.11
     unsus
    0.11
     deceptive
    0.10
     decept
    0.10
     camouflage
    0.10
     embed
    0.10
    Act Density 0.085%

    No Known Activations