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

    Neuronpedia

    INDEX
    Explanations

    matching tiles

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     postpartum
    -0.08
     forfait
    -0.08
     Handlung
    -0.08
    webdriver
    -0.08
     subpoena
    -0.08
     Commentary
    -0.08
     apartado
    -0.08
     соверш
    -0.07
     sweetheart
    -0.07
    _motion
    -0.07
    POSITIVE LOGITS
     matching
    0.11
     Matching
    0.10
     mismatch
    0.10
    matching
    0.09
    Matching
    0.09
     mism
    0.09
    Mismatch
    0.09
     tile
    0.09
    _MATCH
    0.09
    (tile
    0.09
    Act Density 0.014%

    No Known Activations