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

    Neuronpedia

    INDEX
    Explanations

    concrete

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    utoff
    -0.07
    Classification
    -0.07
     pathname
    -0.06
    .lp
    -0.06
    -period
    -0.06
     period
    -0.06
     supervision
    -0.06
     estimation
    -0.06
     Nichols
    -0.06
     sock
    -0.06
    POSITIVE LOGITS
     concrete
    0.13
    Concrete
    0.12
     Concrete
    0.12
     конкрет
    0.08
    奈
    0.08
     tangible
    0.08
    Create
    0.07
    amu
    0.07
     Content
    0.07
     konkrét
    0.07
    Act Density 0.005%

    No Known Activations