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

    Neuronpedia

    INDEX
    Explanations

    demonstrate

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Tell
    0.89
     Tell
    0.78
    Whatever
    0.73
     irgend
    0.67
     ANYTHING
    0.64
     Telling
    0.64
    Anything
    0.62
    tell
    0.62
    “
    0.62
    मग
    0.62
    POSITIVE LOGITS
     demonstrated
    1.62
     demonstrate
    1.53
     demonstrates
    1.51
     demonstration
    1.49
     demost
    1.43
     correctly
    1.43
     demonstrating
    1.42
     explicitly
    1.42
     Demonstr
    1.39
    demon
    1.36
    Act Density 1.794%

    No Known Activations