Llama2 docs added

ai-ml/trainium-inferentia/examples/gradio-ui/gradio-app.py

@@ -21,15 +21,11 @@ def text_generation(message, history):
 
         full_output = response.text
         # Removing the original question from the output
-        answer_only = full_output.replace(prompt, "", 1).strip()
-
-        # Remove any leading/trailing square brackets and double quotes
-        answer_only = answer_only.strip('["]')
+        answer_only = full_output.replace(prompt, "", 1).strip('["]?\n')
 
         # Safety filter to remove harmful or inappropriate content
         answer_only = filter_harmful_content(answer_only)
-
-        return f"<p>{answer_only}</p>"  # Return text with preserved newlines
+        return answer_only
     except requests.exceptions.RequestException as e:
         # Handle any request exceptions (e.g., connection errors)
         return f"AI: Error: {str(e)}"
@@ -46,7 +42,7 @@ def filter_harmful_content(text):
 # Define the Gradio ChatInterface
 chat_interface = gr.ChatInterface(
     text_generation,
-    chatbot=gr.Chatbot(height=400, line_breaks=True),
+    chatbot=gr.Chatbot(line_breaks=True),
     textbox=gr.Textbox(placeholder="Ask me a question", container=False, scale=7),
     title="Llama2 AI Chat",
     description="Ask me any question",

ai-ml/trainium-inferentia/examples/ray-serve/llama2-inf2/ray-service-llama2.yaml

@@ -108,7 +108,7 @@ spec:
 apiVersion: networking.k8s.io/v1
 kind: Ingress
 metadata:
-  name: gradio-ingress
+  name: llama2-ingress
   namespace: llama2
   annotations:
     nginx.ingress.kubernetes.io/rewrite-target: "/$1"

website/docs/gen-ai/index.md

@@ -0,0 +1,19 @@
+---
+sidebar_position: 1
+sidebar_label: Overview
+---
+
+# Gen AI on EKS
+
+Welcome to Gen AI on Amazon Elastic Kubernetes Service (EKS), your gateway to harnessing the power of Large Language Models (LLMs) for a wide range of applications. This introduction page serves as your starting point to explore our offerings for Training, Fine-tuning, and Inference using various LLMs, including BERT-Large, Llama2, Stable Diffusion, and more.
+
+## [Training](https://awslabs.github.io/docs/category/training-on-eks)
+Are you ready to dive into the world of LLMs and train models for your specific needs? Discover our comprehensive Training resources to get started.
+
+## [Fine-tuning](https://awslabs.github.io/docs/category/training-on-eks)
+Fine-tuning LLMs is crucial for tailoring them to your specific tasks. Explore our Fine-tuning section to learn how to adapt LLMs to your unique requirements.
+
+## [Inference](https://awslabs.github.io/docs/category/inference-on-eks)
+Unlock the potential of LLMs for powerful inference tasks. Our Inference resources will guide you through deploying LLMs effectively.
+
+Whether you're an experienced practitioner or new to the field, our Gen AI on EKS capabilities empower you to harness the latest advancements in language modeling. Dive into each section to begin your journey.

website/docs/gen-ai/inference/StableDiffusion.md

@@ -0,0 +1,12 @@
+---
+title: Stable Diffusion on EKS
+sidebar_position: 2
+---
+
+:::info
+
+COMING SOON
+
+Please note that this section is currently a work in progress and will serve as a comprehensive collection of resources for running data and ML workloads on EKS.
+
+:::

website/docs/gen-ai/inference/_category_.json

@@ -0,0 +1,7 @@
+{
+  "label": "Inference on EKS",
+  "position": 1,
+  "link": {
+    "type": "generated-index"
+  }
+}

website/docs/gen-ai/training/BERT-Large.md

@@ -0,0 +1,12 @@
+---
+title: BERT-Large on EKS
+sidebar_position: 1
+---
+
+:::info
+
+COMING SOON
+
+Please note that this section is currently a work in progress and will serve as a comprehensive collection of resources for running data and ML workloads on EKS.
+
+:::

website/docs/gen-ai/training/_category_.json

@@ -0,0 +1,7 @@
+{
+  "label": "Training on EKS",
+  "position": 2,
+  "link": {
+    "type": "generated-index"
+  }
+}

website/docusaurus.config.js

@@ -61,6 +61,12 @@ const config = {
             position: 'left',
             label: 'Introduction',
           },
+          {
+            type: 'doc',
+            docId: 'gen-ai/index',
+            position: 'left',
+            label: 'Gen AI'
+          },
           {
             type: 'doc',
             docId: 'blueprints/amazon-emr-on-eks/index',
@@ -79,12 +85,6 @@ const config = {
             position: 'left',
             label: 'Benchmarks'
           },
-          {
-            type: 'doc',
-            docId: 'workshop/intro',
-            position: 'left',
-            label: 'Workshop'
-          },
           {
             type: 'doc',
             docId: 'resources/intro',

website/sidebars.js

@@ -15,12 +15,11 @@
 const sidebars = {
   // By default, Docusaurus generates a sidebar from the docs folder structure
   // docs: [{type: 'autogenerated', dirName: '.'}],
-
   // But you can create a sidebar manually
+    genai: [{type: 'autogenerated', dirName: 'gen-ai'}],
     blueprints: [{type: 'autogenerated', dirName: 'blueprints'}],
     bestpractices: [{type: 'autogenerated', dirName: 'bestpractices'}],
     benchmarks: [{type: 'autogenerated', dirName: 'benchmarks'}],
-    workshop: [{type: 'autogenerated', dirName: 'workshop'}],
     resources: [{type: 'autogenerated', dirName: 'resources'}],
 };