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

    Neuronpedia

    INDEX
    Explanations

    enclosures/boundaries

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Jord
    -0.09
    ψε
    -0.08
     gloss
    -0.08
    ș
    -0.08
     зі
    -0.07
    🏼
    -0.07
     stelde
    -0.07
    eus
    -0.07
    Installer
    -0.07
    🏻
    -0.07
    POSITIVE LOGITS
     pollution
    0.08
     Pollution
    0.08
     некотор
    0.08
     exatamente
    0.08
     Spill
    0.08
     exactement
    0.07
     exactamente
    0.07
    (pointer
    0.07
    _decay
    0.07
    amatan
    0.07
    Act Density 0.007%

    No Known Activations