Update README.md

README.md

@@ -5,7 +5,7 @@ language: sv
 # A Swedish Bert model
 
 ## Model description
-This model has the same architecture as the Bert Large model in [this paper](https://arxiv.org/abs/1810.04805). It was trained with a batch size of 512 in 600k steps. It is implemented with the Megatron Bert Architecture containing following parameters:
+This model follows the Bert Large model architecture as implemented in [Megatron-LM framework](https://github.com/NVIDIA/Megatron-LM). It was trained with a batch size of 512 in 600k steps. The model contains following parameters:
 <figure>
 
 | Hyperparameter       | Value      |
@@ -18,7 +18,7 @@ This model has the same architecture as the Bert Large model in [this paper](htt
 
 
 ## Training data
-The model is pretrained on a Swedish text corpus of around 80 GB from a variety of sources as shown below.
+The model is pretrained on a Swedish text corpus of around 85 GB from a variety of sources as shown below.
 <figure>
 
 | Dataset       | Genre      | Size(GB)|