Some helpful links regarding training your own LLM since I've been playing with:
https://github.com/geekylink/PicoGPT
Hacker News post: https://news.ycombinator.com/item?id=36832572
LLAMA: https://github.com/ggerganov/llama.cpp
Dead simple LLAMA: https://cocktailpeanut.github.io/dalai/#/
Training data:
OpenWebText2 is an enhanced version of the original OpenWebTextCorpus covering all Reddit submissions from 2005 up until April 2020, with further months becoming available after the corresponding PushShift dump files are released.
https://openwebtext2.readthedocs.io/en/latest/