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

    Neuronpedia

    INDEX
    Explanations

    research methodologies

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     techniques
    -1.00
     certainly
    -0.93
     technique
    -0.80
    certainly
    -0.68
    Thus
    -0.68
     Certainly
    -0.67
     Marks
    -0.66
     Thus
    -0.65
    Certainly
    -0.59
     marks
    -0.57
    POSITIVE LOGITS
     jurídica
    0.78
     religieuse
    0.77
     religieuses
    0.77
     TextInputType
    0.74
     vœux
    0.72
     dislocations
    0.72
     financières
    0.71
     débats
    0.71
     morada
    0.69
     silvestre
    0.69
    Act Density 0.141%

    No Known Activations