-
Howdy!
I am using the `finetune/finetune.py` script. It trains on NVIDIA A40, and at the end when it tries to save the model/checkpoints it raises the `torch.cuda.OutOfMemoryError: CUDA out of mem…
-
There are open source implementation and resources available for training QA models. We should collect and discuss those resources here. Once the discussion converges, we should use the frameworks for…
-
# Trending repositories for C#
1. [**OpenRA / OpenRA**](https://github.com/OpenRA/OpenRA)
__Open Source real-time strategy game engine for early Westwood games such as Command & C…
-
We could add the prompt retrieval benchmark: https://arxiv.org/abs/2209.01975
-
I would like to request a new feature that allows users to send requests to a custom GPT model instead of the default new chat model.
This feature would enable more customization and flexibility, ie…
-
## Idea Description:
The app should have an NPC Builder function that allows the user to create new NPCs with a background story, name, and game-relevant stats. The user should be able to specify the…
-
-
### Checked other resources
- [X] I added a very descriptive title to this issue.
- [X] I searched the LangChain.js documentation with the integrated search.
- [X] I used the GitHub search to fin…
-
All files, optimize the GPT token usage so that it doesn't go through them as fast. Create a MD file named updates and log all updates and tell me what they are and how they better optimize it
-
I really like your plugin, it really helps me a lot.
But in the process of using it, I chose a deck with a lot of sub-decks, but the plugin would indicate that no card could be found. I included a lo…