@ChrisChipMonk (Self-Correction:) The earlier DeepSeek v3 and even prior generations of DeepSeek LLMs had a similar hybrid-MoE arch. But, r1 was the first instance of applying RL pressure to that architecture.
in reply to: 1917741724308496872
same thread: 1917687241666679170 1917692380431671586 1917738496824926458 1917740605825855579 1917740903558701087 1917741724308496872 1917865196548506065 1917866498275582343 1917874329192145181 1917887008174723138 1917888335441489926 1917889406268068102 1917903929914126808 1917909202766352801
cited on: deepseek-v3
Reproduced against link rot, credited and linked to its original. Yours and you’d rather it weren’t here? Open an issue.