lit-llama
https://github.com/lightning-ai/lit-llama
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
not yet supported0 Subscribers
Add a CodeTriage badge to lit-llama
Help out
- Issues
- Parallelize prepare_redpajama.py through multiprocessing
- Add script to train tokenizer on RedPajama
- Using lit-llama for multitask learning
- Use all gpus available for training
- `open in google colab` integration
- `generate.py` and `generate/full.py` could be merged
- cublasLt runs into an error on 8 bit quantized
- Less is more for alignment (LIMA) - adding special EOT token
- explanation of the hyperparameters in the pretraining script
- Convert lit-llama weights to huggingface
- Docs
- not yet supported