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

    Neuronpedia

    INDEX
    Explanations

    recognises or utilises descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     şi
    2.25
     flavorful
    2.05
     neighborhoods
    2.02
     lackluster
    1.96
     demeanor
    1.95
    Neighbors
    1.89
     grayish
    1.88
     odors
    1.82
    Agregar
    1.82
     laborers
    1.82
    POSITIVE LOGITS
     فى
    2.95
    Whilst
    2.92
     Whilst
    2.78
     whilst
    2.61
     utilising
    2.54
     recognised
    2.49
     maximising
    2.47
     recognisable
    2.46
     recognise
    2.46
     recognises
    2.46
    Act Density 0.006%

    No Known Activations