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

    Neuronpedia

    INDEX
    Explanations

    lacking or meeting requirements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ４
    0.44
    7
    0.44
    4
    0.44
    ７
    0.39
    5
    0.38
    6
    0.36
    8
    0.36
    ５
    0.36
    ８
    0.36
    <0x9F>
    0.36
    POSITIVE LOGITS
     aforementioned
    0.41
     shenanigans
    0.38
     nuanced
    0.37
     matchup
    0.37
     disequ
    0.37
     antics
    0.36
     unorthodox
    0.36
     مذکور
    0.36
     dodgy
    0.35
     ulterior
    0.35
    Act Density 0.789%

    No Known Activations