WIP

Author: adam-cowley

asciidoc/courses/llm-chatbot-typescript/modules/1-project-setup/lessons/2-setup/lesson.adoc

@@ -84,17 +84,30 @@ NEO4J_USERNAME="{sandbox_username}"
 NEO4J_PASSWORD="{sandbox_password}"
 ----
 
-You must also set your OpenAI API key.
-You can find link:https://help.openai.com/en/articles/4936850-where-do-i-find-my-openai-api-key[instructions on generating an API key on openai.com^].
+You will also need an API Key to generate LLM responses using OpenAI.
+
+We have generated an OpenAI API key for you to use through our OpenAI Proxy for the duration of this course using a proxy.
+The API key will be limited to 5 requests every two minutes.
+
 
 Replace the `OPENAI_API_KEY` value with your OpenAI API key.
 
 .env.local
-[source,env]
+[source,env,subs="attributes+"]
 ----
-OPENAI_API_KEY="sk-..."
+OPENAI_API_KEY="{llm-api-key}"
+OPENAI_API_BASE="{llm-api-base}"
 ----
 
+If you find the rate limits restricting, you can always use an OpenAI API Key.
+In this case, you would omit the Base URL.
+You can find link:https://help.openai.com/en/articles/4936850-where-do-i-find-my-openai-api-key[instructions on generating an API key on openai.com^].
+
+This course's challenges do not require you use OpenAI LLMs, so feel free to use an alternative LLM provider.
+We also like link:https://ollama.com/[Ollama^].
+
+
+
 [WARNING]
 .Keep your secrets safe
 ====
@@ -151,6 +164,6 @@ include::questions/1-server.adoc[leveloffset=+1]
 [.summary]
 == Summary
 
-In this lesson, you obtained a copy of the course code, installed the dependency and used the `npm run dev` command to start the app.
+In this lesson, you obtained a copy of the course code, installed the dependency, and used the `npm run dev` command to start the app.
 
 In the next lesson, you will set the scope for the project.

asciidoc/courses/llm-chatbot-typescript/modules/2-chains/lessons/1-introduction-to-lcel/images/chain.png

@@ -26,6 +26,8 @@ You can link:https://js.langchain.com/docs/expression_language/[read more about
 In the link:https://graphacademy.neo4j.com/courses/llm-fundamentals/3-intro-to-langchain/2.5-chains/[Chains lesson in Neo4j & LLM Fundamentals^], you learned about the `LLMChain`.
 The link:https://api.python.langchain.com/en/latest/_modules/langchain/chains/llm.html#LLMChain[`LLMChain`^] is an example of a simple chain that, when invoked, takes a user input, replaces the value inside the prompt and passes the prompt to an LLM and specifies the result.
 
+image::./images/chain.png[Prompt → LLM → Response]
+
 // [source]
 // ----
 // // TODO: Diagram
@@ -60,6 +62,16 @@ The prompt will be passed to an LLM, in this case, the `ChatOpenAI` model.
 include::{repository-raw}/main/examples/chain.mjs[tag=llm]
 ----
 
+=== The Response Parser
+
+You will then pass the LLM's response to an output parser.
+The most straightforward parser is a `StringOutputParser`.
+
+[source,typescript]
+----
+include::{repository-raw}/main/examples/chain.mjs[tag=parser]
+----
+
 
 === Creating the Chain
 

asciidoc/courses/llm-chatbot-typescript/modules/2-chains/lessons/1-introduction-to-lcel/lesson.adoc

@@ -19,7 +19,7 @@ The agent currently uses a pre-written prompt from LangChain Hub.
 You can link:https://smith.langchain.com/hub/hwchase17/openai-functions-agent[view this prompt in LangChain Hub^].
 
 The prompt is relatively basic.
-It consists of an array of messages consisting of a role definition, human input, and placeholders.
+It consists of an array of messages, including a role definition, human input, and placeholders.
 
 [source]
 .The prompt
@@ -47,7 +47,7 @@ For example, the following prompt instructs the LLM to refuse to answer question
 .Modified Prompt
 [source]
 ----
-include::{repository-raw}/main/src/{lab-solution}[tag="scoped", indent=0]
+include::{repository-raw}/main/{lab-solution}[tag="scoped", indent=0]
 ----
 
 [NOTE]
@@ -65,7 +65,7 @@ The link:{repository-blob}/main/{test-file}[unit test^] will verify whether the
 == Experiment
 
 You can extend the prompt to instruct the agent to act in specific ways.
-Experiment with different prompts to see how the agent responds to various types of questions.
+Experiment with different prompts to see how the agent responds to various questions.
 
 When you are ready, click Continue.
 

asciidoc/courses/llm-chatbot-typescript/modules/6-agent/lessons/6-limiting-scope/lesson.adoc

@@ -20,37 +20,7 @@ This course includes instructions for using link:https://openai.com/[OpenAI^], b
 
 [NOTE]
 .Using OpenAI
-====
-We have generated an OpenAI API key for you to use through our OpenAI Proxy for the duration of this course using a proxy.
-The API key will be limited to 5 requests every two minutes.
-
-
-.Environment Variable
-[source,env,subs="attributes+"]
-----
-OPENAI_API_KEY={llm-api-key}
-----
-
-You must also set the `base_url` parameter to use our proxy server.
-
-.Setting the Proxy
-[source,python,subs="attributes+"]
-----
-from openai import OpenAI
-
-model = OpenAI(
-  api_key="{llm-api-key}",
-  base_url="https://graphacademy.neo4j.com/api/llm/v1/"
-)
-----
-
-You can always use an existing OpenAI API key and omit the `base_url` argument.
-
-
-====
-
-
-
+If you wish to use OpenAI and follow this course's practical activities, you must create an account and set up billing.
 
 == Setup
 
@@ -84,7 +54,7 @@ pip install openai langchain-openai
 
 == Create a Langchain application
 
-Create a new Python program and copy this code into a new Python file.
+Create a new Python program and copy this code into it.
 
 [source,python]
 ----

asciidoc/courses/llm-fundamentals/modules/3-intro-to-langchain/lessons/2-initialising-the-llm/lesson.adoc

@@ -2,7 +2,9 @@
 from langchain.prompts import PromptTemplate
 from langchain.chains import LLMChain
 
-llm = OpenAI(openai_api_key="sk-...")
+llm = OpenAI(
+  openai_api_key="sk-..."
+)
 
 template = PromptTemplate.from_template("""
 You are a cockney fruit and vegetable seller.
@@ -13,8 +15,8 @@
 """)
 
 llm_chain = LLMChain(
-    llm=llm,
-    prompt=template
+  llm=llm,
+  prompt=template
 )
 
 response = llm_chain.invoke({"fruit": "apple"})

asciidoc/courses/llm-fundamentals/modules/3-intro-to-langchain/lessons/2.5-chains/code/llm_chain.py

@@ -4,7 +4,8 @@
 from langchain.schema import StrOutputParser
 
 llm = OpenAI(
-    openai_api_key="sk-...")
+  openai_api_key="sk-..."
+)
 
 template = PromptTemplate.from_template("""
 You are a cockney fruit and vegetable seller.
@@ -15,9 +16,9 @@
 """)
 
 llm_chain = LLMChain(
-    llm=llm,
-    prompt=template,
-    output_parser=StrOutputParser()
+  llm=llm,
+  prompt=template,
+  output_parser=StrOutputParser()
 )
 
 response = llm_chain.invoke({"fruit": "apple"})

asciidoc/courses/llm-fundamentals/modules/3-intro-to-langchain/lessons/2.5-chains/code/llm_chain_output.py

@@ -4,7 +4,8 @@
 from langchain.output_parsers.json import SimpleJsonOutputParser
 
 llm = OpenAI(
-    openai_api_key="sk-...")
+  openai_api_key="sk-..."
+)
 
 template = PromptTemplate.from_template("""
 You are a cockney fruit and vegetable seller.
@@ -17,9 +18,9 @@
 """)
 
 llm_chain = LLMChain(
-    llm=llm,
-    prompt=template,
-    output_parser=SimpleJsonOutputParser()
+  llm=llm,
+  prompt=template,
+  output_parser=SimpleJsonOutputParser()
 )
 
 response = llm_chain.invoke({"fruit": "apple"})

asciidoc/courses/llm-fundamentals/modules/3-intro-to-langchain/lessons/2.5-chains/code/llm_chain_output_json.py

@@ -8,22 +8,23 @@ Chains allows you to combine language models with different data sources and thi
 
 == Using LLMChain
 
-The simplest chain is an `LLMChain`. An `LLMChain` combines a prompt template with an LLM and returns a response.
+The most straightforward chain is an `LLMChain`. An `LLMChain` combines a prompt template with an LLM and returns a response.
 
 Previously, you created a program that used a prompt template and an LLM to generate a response about fruit.
 
+
 [%collapsible]
 .Click to reveal the code for the program.
 ====
-[source,python]
+[source,python,subs="attributes+"]
 ----
 include::../2-initialising-the-llm/code/llm_prompt.py[]
 ----
 ====
 
 You can combine this program into a chain and create a reusable component.
 
-[source,python]
+[source,python,subs="attributes+"]
 ----
 include::code/llm_chain.py[]
 ----