README updated with example conversation

README.md

@@ -18,18 +18,20 @@ DialoGPT_SV
 
 This is a fine-tuned model of the DialoGPT (medium) on the Swedish Gothenburg Dialogue Corpus (GDC). It is intended to be used as a Swedish conversational system. The GDC dataset it's trained on is limited in scope, as it's from the transcription of dialogues of about 25 different social activities, including debates, academic seminars and situations. The dataset has about 108K turns. 
 
-The perplexity achieved on the test and validation sets (from 80:10:10 split) are 27.81 and 27.81, respectivley. More information about the original pre-trained model can be found [here] (https://huggingface.co/microsoft/DialoGPT-medium)
+The perplexity achieved on the test and validation sets (from 80:10:10 split) are 27.81 and 27.81, respectivley. More information about the original pre-trained model can be found [here](https://huggingface.co/microsoft/DialoGPT-medium)
 
 * Generation example from an interactive environment:
 
 |Role | Response |
 |---------|------------|
-|User | |
-|Bot | |
+|User | Jag ska fika |
+|Bot | ja det är ju så att jag kanske inte kan få den att göra det |
+|User | tack så mycket |
+|Bot | tack tack ses vi om sex vecko igen | 
 
 Please find the information about preprocessing, training and full details of the DialoGPT in the [original DialoGPT repository](https://github.com/microsoft/DialoGPT)
 
-The paper for this work was accepted at the Northern Lights Deep Learning conference 2022. Arxiv paper: [https://arxiv.org/pdf/2110.06273.pdf] (https://arxiv.org/pdf/2110.06273.pdf)
+The paper for this work was accepted at the Northern Lights Deep Learning conference 2022. Arxiv paper: [https://arxiv.org/pdf/2110.06273.pdf](https://arxiv.org/pdf/2110.06273.pdf)
 
 ### How to use