This repo contains multiple codes related to the usage of LLM on various problems like question answering, text summarization, etc.
- SOTA models like LLama2, DeBERTa used in the codes.
- RAG creation from wiki data shown in the codes. There are multiple ways to create RAG either via the technique shown in the codes, via llamaindex, etc.
- Used Faiss for indexing embeddings in RAG. One can also use Pinecone, etc for indexing.
- Finetuning LLM via PEFT techniques is also shown on text summarization.
- Do watch out this space for more informative codes and share your feedback if these codes are useful.