update the description

app.py

@@ -1,20 +1,7 @@
 import os
-# import threading
 import streamlit as st
 from itertools import tee
 from model import InferenceBuilder
-# from chain import ChainBuilder
-
-# DATABRICKS_HOST = os.environ.get("DATABRICKS_HOST")
-# DATABRICKS_TOKEN = os.environ.get("DATABRICKS_TOKEN")
-# remove these secrets from the container
-# VS_ENDPOINT_NAME = os.environ.get("VS_ENDPOINT_NAME")
-# VS_INDEX_NAME = os.environ.get("VS_INDEX_NAME")
-
-# if DATABRICKS_HOST is None:
-#     raise ValueError("DATABRICKS_HOST environment variable must be set")
-# if DATABRICKS_TOKEN is None:
-#     raise ValueError("DATABRICKS_TOKEN environment variable must be set")
 
 MODEL_AVATAR_URL= "./iphone_robot.png"
 MAX_CHAT_TURNS = 10 # limit this for preliminary testing
@@ -33,10 +20,11 @@ EXAMPLE_PROMPTS = [
 TITLE = "CyberSolve LinAlg 1.2"
 DESCRIPTION= """Welcome to the CyberSolve LinAlg 1.2 demo! \n
 
-**Overview and Usage**: This 🤗 Space is designed to demo the abilities of the CyberSolve LinAlg 1.2 text-to-text language model.
-and is augmented with additional organization-specific knowledge. Particularly, it has been preliminarily augmented with knowledge of Vanderbilt University Medical Center
-terms like **EDW**, **HCERA**, **NRHA** and **thousands more**. (Ask the assistant if you don't know what any of these terms mean!) On the left is a sidebar of **Examples**; 
-click any of these examples to issue the corresponding query to the AI.
+**Overview and Usage**: This 🤗 Space is designed to demo the abilities of the **CyberSolve LinAlg 1.2** text-to-text language model. Specifically, the CyberSolve LinAlg 1.* family of models
+are downstream versions of the 783M parameter FLAN-T5 text-to-text transformer, fine-tuned on the Google DeepMind Mathematics dataset for the purpose of solving linear equations of a single variable. 
+To effectively query the model for its intended task, prompt the model solve an arbitrary linear equation of a single variable with a query of the form: "Solve 24 = 1601*c - 1605*c for c."; the model 
+will return its prediciton in a simple format. The algebraic capabailites far exceed those of the base FLAN-T5 model. CyberSolve LinAlg 1.2 achieves a 90.7 percent exact match benchmark on the DeepMind Mathematics
+evaluation dataset of 10,000 unique linear equations; the FLAN-T5 base model scores 9.6 percent. On the left is a sidebar of **Examples** that can be clicked to query to model.
 
 **Feedback**: Feedback is welcomed, encouraged, and invaluable! To give feedback in regards to one of the model's responses, click the **Give Feedback on Last Response** button just below
 the user input bar. This allows you to provide either positive or negative feedback in regards to the model's most recent response. A **Feedback Form** will appear above the model's title.