The video showcases the installation of Llama 2 7B LLM in 1-bit precision by Mobius Labs. The model was quantized using HQQ and fine-tuned with LoRA on 25k samples. The content also includes links to become a Patron, as well as the creator’s LinkedIn, YouTube, and blog. Related videos on using custom datasets with Mixtral 8x7B and an introduction to AWS Bedrock are also mentioned. The video is copyrighted to Fahd Mirza in 2021.
Source link
Source link: https://www.youtube.com/watch?v=Kp48oVnJFxw
LLM Installation for 1-Bit Precision Llama 2 7B #Llama
