/LLM-playground

Primary LanguageJupyter Notebook

LLM Playground

Instruction-Tuning

  • Model: Llama-2 (NousResearch/Llama-2-7b-chat-hf)
  • Dataset: mlabonne/guanaco-llama2-1k
  • Finetuning Llama-2 using QLoRA to improve performance in instruction-following tasks.