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

    Neuronpedia

    INDEX
    Explanations

    negative concepts

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    days
    -0.08
    write
    -0.07
    datap
    -0.06
     yp
    -0.06
     _.
    -0.06
    cou
    -0.06
    terms
    -0.06
    forEach
    -0.06
    Barrier
    -0.06
    getTable
    -0.06
    POSITIVE LOGITS
     aston
    0.06
     costing
    0.06
     madrid
    0.06
     quali
    0.06
     Costs
    0.06
    (layer
    0.06
    _IT
    0.05
    tr
    0.05
     muschi
    0.05
    ossil
    0.05
    Act Density 0.054%

    No Known Activations