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

    Neuronpedia

    INDEX
    Explanations

    extension

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Soda
    -0.07
     breathing
    -0.07
     laying
    -0.07
     Deo
    -0.07
     swallowing
    -0.07
     dormir
    -0.07
    Mario
    -0.07
     nella
    -0.07
     massive
    -0.07
     glycol
    -0.07
    POSITIVE LOGITS
     dissemination
    0.11
     Beratung
    0.11
    beratung
    0.11
     Disse
    0.10
     advisory
    0.10
     консульт
    0.10
     Advis
    0.10
     advising
    0.10
     rådg
    0.10
     beraten
    0.10
    Act Density 0.023%

    No Known Activations