metadata
base_model: meta-llama/Meta-Llama-3-8B-Instruct
library_name: peft
This repository contains the LoRA adapter weights from the fine-tuning of the Llama 3 (8B) model on patent documents using masked next token prediction (MNTP). MNTP is the first step in the adaptation of the base model for embedding generation following the llm2vec approach.
Framework versions
- PEFT 0.12.0