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

    Neuronpedia

    INDEX
    Explanations

    keycaps

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     wire
    0.38
     wires
    0.38
     Wire
    0.38
    Wire
    0.38
    Wires
    0.37
     wired
    0.36
     मक
    0.36
    มอ
    0.36
     지구
    0.36
    Gina
    0.36
    POSITIVE LOGITS
    drawn
    0.47
     drawn
    0.46
    祜
    0.46
    cap
    0.44
     cap
    0.44
     hold
    0.44
    hold
    0.43
     Drawn
    0.42
     Hold
    0.41
     set
    0.41
    Act Density 0.007%

    No Known Activations