winglian

/

llama-3-1m-context-gradient-lora

like 12

Could you provide details on how the LoRA extraction was performed given the Llama-3-8B-Instruct-Gradient-1048k model? Any resources, tools, or insights into the process would be greatly appreciated!