LLaMA2 fine-tuning and instruction training
11 repos
Fine-tuning and instruction-following optimization techniques for LLaMA2 language models across multiple scales (7B, 13B parameters). This cluster covers LoRA-based parameter-efficient training methods, Chinese language adaptation, and chat model development. The repositories provide both base model implementations and specialized variants (Firefly-LLaMA2) demonstrating practical approaches to adapting large language models for specific domains and instruction-following tasks.
Python · 5
C · 1
HTML · 1
bloom ·8,710
llama ·8,698
lora ·8,686
instruction-set ·8,271
instruct-gpt ·8,271
chinese-nlp ·8,271
gpt-evaluation ·8,271
gpt-q ·8,271
instruct-finetune ·8,271
open-models ·8,271
