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

    Neuronpedia

    INDEX
    Explanations

    fully fenced

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Open
    0.75
     चेयर
    0.74
     ओपन
    0.70
    Open
    0.69
     Text
    0.68
    Released
    0.68
    தய
    0.67
     rilas
    0.67
    open
    0.67
    মুখে
    0.65
    POSITIVE LOGITS
    Fully
    0.85
     fully
    0.85
     Fully
    0.78
    RetValue
    0.74
     fenced
    0.70
    搓
    0.70
     entièrement
    0.69
     lớ
    0.69
     livello
    0.67
    FULLY
    0.67
    Act Density 0.044%

    No Known Activations