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

    Neuronpedia

    INDEX
    Explanations

    Neural network training/fine-tuning

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     fulfillment
    -0.08
     portray
    -0.08
     प्रद
    -0.08
     cú
    -0.08
     acute
    -0.08
     गर्�
    -0.08
    -0.07
     REF
    -0.07
     portrays
    -0.07
     оформления
    -0.07
    POSITIVE LOGITS
     doucement
    0.10
     knobs
    0.10
     langsam
    0.09
     Slowly
    0.09
    哪些
    0.09
     lentamente
    0.09
     gradually
    0.09
    GPU
    0.09
     langzaam
    0.09
     Updating
    0.09
    Act Density 0.003%
    No Known Activations