Stories by genji970

Show HN: Way to divide parameter space in LLM training

Applying zigzag algebra in abstract algebra to improve LLM training

Fixing hallucination in LLM prediction with only one 48gib GPU

Show HN: Fine tuning and RLHF mistralai 7B using DeepSpeed

Show HN: RLHF and Lora finetuning to mistralai 7B with DeepSpeed learning

Show HN: Mistralai-7B distributed learning using DeepSpeed pipeline

Show HN: Vlm in 3D PC, 16 shot scanobjectnn top1 acc: 99.91

Show HN: I achieved over 10% improvement on 3D vision PointCLIP