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

    Neuronpedia

    INDEX
    Explanations

    rivers flowing and their origins

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     rivers
    0.85
    河流
    0.84
     реки
    0.75
    River
    0.75
     river
    0.73
     fiume
    0.71
     Rivers
    0.69
    rivers
    0.68
     RIVER
    0.68
     River
    0.64
    POSITIVE LOGITS
     downstream
    0.68
     seawater
    0.55
     sea
    0.53
     contribut
    0.50
     upstream
    0.50
     contrib
    0.49
     recibir
    0.48
     merge
    0.48
     البحر
    0.48
     merging
    0.47
    Act Density 0.019%

    No Known Activations