NousResearch/Meta-Llama-3.1-8B-Instruct Korean finetuned model with SFT->RLHF->DPO

Tools 8B 70B

168 Pulls Updated 5 weeks ago

033465c927a9 · 110B
{ "num_keep": 10, "stop": [ "<|start_header_id|>", "<|end_header_id|>", "<|eot_id|>" ] }