joennlae | Dark Hacker News

1.	Different Sampling Strategies for LLMs(huggingface.co) 1 points by joennlae 2 years ago \| 0 comments
2.	Show HN: Stella Nera – Maddness Hardware Accelerator(github.com) 154 points by joennlae 2 years ago \| 26 comments
3.	Show HN: Maddness – Energy-efficient DNNs without multiplying(arxiv.org) 2 points by joennlae 2 years ago \| 0 comments
4.	Show HN: less than 650 LOC trainable GPT only using NumPy(github.com) 90 points by joennlae 2 years ago \| 18 comments
5.	Show HN: Tensorli – NumPy Only Transformer Training (<650 lines)(github.com) 2 points by joennlae 2 years ago \| 0 comments
6.	GithubNext: AI-Powered Auto-Fix Pull Requests for Issues(githubnext.com) 3 points by joennlae 2 years ago \| 0 comments
7.	Show HN: Tensorli - GPT-Like Transformers in Pure NumPy(github.com) 3 points by joennlae 2 years ago \| 0 comments
8.	Nvidia uses internal LLM to support chip design efforts(blogs.nvidia.com) 2 points by joennlae 2 years ago \| 0 comments
9.	“THIS-IS-NOT-A-BACKUP” forking popular FOSS repos with a cron job(github.com) 1 points by joennlae 4 years ago \| 0 comments