📝 add details on csearch

Signed-off-by: peter szemraj <[email protected]>

README.md

@@ -6,8 +6,9 @@ colorTo: blue
 sdk: gradio
 sdk_version: 3.1.4
 app_file: app.py
-pinned: false
+pinned: true
 license: apache-2.0
+tags: [email, autocomplete, text generation, contrastive search]
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+Check out the configuration reference at <https://huggingface.co/docs/hub/spaces-config-reference>

app.py

@@ -214,7 +214,7 @@ if __name__ == "__main__":
             gr.Markdown("---")
             gr.Markdown("## Advanced Options")
             gr.Markdown(
-                "This demo generates text via beam search. See details about these parameters [here](https://huggingface.co/blog/how-to-generate), otherwise they should be fine as-is."
+                "This demo generates text via the new [constrastive search](https://huggingface.co/blog/introducing-csearch). See details on the csearch blog post for the methods' parameters or [here](https://huggingface.co/blog/how-to-generate), for general decoding."
             )
             with gr.Row():
                 model_name = gr.Dropdown(
@@ -258,7 +258,7 @@ if __name__ == "__main__":
 
             gr.Markdown("## About")
             gr.Markdown(
-                "[This model](https://huggingface.co/postbot/distilgpt2-emailgen) is a fine-tuned version of distilgpt2 on a dataset of 50k emails sourced from the internet, including the classic `aeslc` dataset.\n\nCheck out the model card for details on notebook & command line usage."
+                "[This model](https://huggingface.co/postbot/distilgpt2-emailgen) is a fine-tuned version of distilgpt2 on a dataset of 100k emails sourced from the internet, including the classic `aeslc` dataset.\n\nCheck out the model card for details on notebook & command line usage."
             )
             gr.Markdown(
                 "The intended use of this model is to provide suggestions to _auto-complete_ the rest of your email. Said another way, it should serve as a **tool to write predictable emails faster**. It is not intended to write entire emails from scratch; at least **some input** is required to guide the direction of the model.\n\nPlease verify any suggestions by the model for A) False claims and B) negation statements **before** accepting/sending something."