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

    Neuronpedia

    INDEX
    Explanations

    knobs and dials

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     vented
    0.85
     وت
    0.80
     takes
    0.79
     تت
    0.75
     م
    0.75
    сут
    0.72
     sells
    0.72
     backed
    0.71
     took
    0.71
     hitched
    0.70
    POSITIVE LOGITS
     Civ
    0.95
     Attorney
    0.94
     Matter
    0.92
     Guard
    0.91
     Policy
    0.89
    Law
    0.89
     Instruction
    0.88
     File
    0.88
     Peace
    0.88
     Bag
    0.88
    Act Density 0.011%

    No Known Activations