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

    Neuronpedia

    INDEX
    Explanations

    light weight

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     obligations
    -0.07
    Eight
    -0.07
    Vis
    -0.06
    ith
    -0.06
    iks
    -0.06
     bishops
    -0.06
     enr
    -0.06
    ominated
    -0.06
     fingertips
    -0.05
    .im
    -0.05
    POSITIVE LOGITS
     Packaging
    0.08
    ************************
    0.07
     McCart
    0.07
     lightweight
    0.07
     IData
    0.07
    碎
    0.07
     руч
    0.07
    ества
    0.07
    734
    0.06
    =plt
    0.06
    Act Density 0.011%

    No Known Activations