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

    Neuronpedia

    INDEX
    Explanations

    complexity of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     semplic
    0.77
    Tc
    0.62
     absence
    0.61
     sencillo
    0.61
     ngắn
    0.61
     eenvoudig
    0.60
    absence
    0.58
     ignorance
    0.57
     ignorant
    0.57
     ineffective
    0.57
    POSITIVE LOGITS
     larger
    1.21
    更高
    1.21
    更大
    1.16
     bigger
    1.13
    更大的
    1.13
     größere
    1.08
     større
    1.05
     beyond
    1.04
    更高的
    1.04
    larger
    1.03
    Act Density 2.508%

    No Known Activations