[markrubin]

I managed to get GPT4All running on my 2021 MBP M1. I did a bunch of stuff to troubleshoot this: uninstalled, reinstalled, recompiled from source, updated all related tools, etc.  In the end, the problem was with the LocalDocs plugin.

I accidentally indexed a giant Google Drive folder structure (about 200 GB). This had a big performance hit.  It took 20 to 30 min to get a reply and it crashed most of the time.

The maintainer uninstall process doesn’t reset the pointers to the file and I couldn’t uninstall it from the UI because it crashed during the process.  I figured out that the local index is stored in the same folder as the models.  I manual deleted it and restarted GPT4All… it’s working fine now.  Performance is fine for doing the Save Wisdom project.

https://savewisdom.org/the-1000-word-save-wisdom-questions/

[markrubin]

1. Fostering self-awareness: How much time do I sit with my feelings before I make up a story about the reason for those feelings?  When I reflect on the story, do I consider all perspectives?  Do I respond with empathy for those perspectives?  Does this story bring me into connection or isolation with the other perspectives?  How can I improve my self-awareness as a result of this story?
2. Cultivating wisdom: How does this new self-awareness impact my sense of Intuition? Can the new perspectives I considered above improve my understanding of other stories in new ways?  Does this give me any new insights, vision, or clarity of any present situations?
3. Appreciating personal growth: Can I extract wisdom from new knowledge by asking “why” until the answer matches the two most important variables?  How can I apply this why to my values and pursuits of meaning and purpose to develop character?
4. Identifying blind spots: What have I learned that surprised me?  What blind spot did this reveal?  Can I learn more about this area to improve?

Attachments:

You must be logged in to view attached files.

[markrubin]

Hey all,

I’d love to get GPT4All running on my 2021 M1 Macbook Pro.  I only have 16 GB of RAM.  It takes 20-30 min to get a response.  After the first response, the chat interface crashes. I have a feeling it is a config issue, but I can’t find it.  When I click “Stop generating” the there is no response from the UI so I think there’s a problem between the chat interface and the rest of it.

I uninstalled and reinstalled twice and I’m still not having any luck.  I tried to build it from the source code, but there are some dependencies I can’t sort out.

Does anyone know if the uninstall tool actually uninstalls all of the files?  Or does anyone know if there’s a way to manually uninstall everything and start over?  I looked at the GPT4All site and couldn’t find any support of forums so I thought I’d try here.

Oh, my Mac has a plain vanilla setup… no custom stuff.  Thanks!

 

[markrubin]

I’ve got a 2020 M1 MBP with 16 Gb of RAM and it takes about 20 minutes to get a reply from the GPT4all prompt.  I can’t tell if there’s a problem with the installation or if I’m constrained on RAM.  Does anyone have it running on a similar spec machine?

[markrubin]

Hi, I’m Mark. I’m interested in the convergence of convergences and the cycles within cycles.  Also, I enjoy creating the future that I imagine and I feel very fortunate to have discovered the tools, prompts, and ideas on Multiplex.  I’m launching an eco project now and the content here is helping me optimize some things.

I’m on Twitter at:  https://twitter.com/markianrubin

I’m happy to connect and share ideas.

 

 

[Author]

Posts