Skip to content

Commit

Permalink
Update 7-local-genai-intro.md
Browse files Browse the repository at this point in the history
  • Loading branch information
richmccue authored Oct 29, 2024
1 parent 5378af5 commit 3a74f69
Showing 1 changed file with 4 additions and 4 deletions.
8 changes: 4 additions & 4 deletions 7-local-genai-intro.md
Original file line number Diff line number Diff line change
Expand Up @@ -63,18 +63,18 @@ Now we have _GPT4All_ installed on your laptop, we need to install a large langu

## Install and Test another Language Model
Let's try installing a second Language Model so that we can compare the results of the questions you've asked the _Llama 3 Instruct_ model with another model:
<img src="images/7-other-models.png" style="width:8000px;" alt="Download new Models">
1. On the left navigation bar click on the green **Models** button.
2. Click on the gree **Add Model** button on the top right of the screen.
3. Scroll to find the **Nous Hermes 2 Mistral DPO** model, and then click on the green **Download** button to the right of the model description.
<img src="images/7-other-models.png" style="width:8000px;" alt="Download new Models">
4. Scroll to find the **Nous Hermes 2 Mistral DPO** model, and then click on the green **Download** button to the right of the model description.
> _Note: The download is quite large at 4.3GB so it may take a while for it to download. For this model, GPT4All recommends that your computer have at least 8GB of memory for your computer to run the model at a reasonable speed._ If you have any questions about this please ask your instructor.
4. Once the model has finished downloading, click on the green **X** in the top right to close the "Discover and Download Models" window.
5. Before we can start testing the new model, we need to tell GPT4All that we want to use the _Nous Hermes 2 Mistral DPO_ model:
- Click on the **Choose a model...** drop down box at the middle top of the window.
- Select **Nous Hermes 2 Mistral DPO** model. Depending on the speed of your laptop it should take between take 10 to 60 seconds for the model to finish loading.
- Scroll down a ways and select **Nous Hermes 2 Mistral DPO** model. Depending on the speed of your laptop it should take between take 10 to 60 seconds for the model to finish loading.
3. Let's test _GPT4All & Nous Hermes 2 Mistral DPO_ with a generic prompt to start with. Copy and paste the following text into the Send a Message field and review the results:<br>
```Why would someone want to attend the University of Victoria in Canada? Why might the university not be a good fit for someone? Provide citations.```<br>
<img src="images/7-first-prompt.png" style="width:800px;" alt="First prompt">
<img src="images/7-first-prompt2.png" style="width:800px;" alt="First prompt">
- How reasonable do the results look to you?
- Do you notice any significant omissions from the "not be a good fit" section?
- Are there any major differences from the _Llama 3 Instruct_ response? Is there any mention of **Cost of living or Housing availability**?
Expand Down

0 comments on commit 3a74f69

Please sign in to comment.