Don't be lazy: CompleteP enables compute-efficient deep transformers · HackerPapers discussionshttps://hackerpaper.kaipfister.de/p/2505.01618/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMTDon't be lazy: CompleteP enables compute-efficient deep transformers — mentioned by yorwbahttps://news.ycombinator.com/item?id=44904270urn:hackerpapers:mention:2505.01618:44904270Thu, 14 Aug 2025 19:01:57 GMT<p>Doing hyperparameter sweeps on lots of small models to find the optimal values for each size and fitting scaling laws to predict the hyperparameters to use for larger models seems to work reasonably well. I think https://arxiv.org/abs/2505.01618 is the latest advance in that vein.</p><p><a href="https://hackerpaper.kaipfister.de/p/2505.01618/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.LGDon't be lazy: CompleteP enables compute-efficient deep transformers — HN submissionhttps://news.ycombinator.com/item?id=43906014urn:hackerpapers:submission:2505.01618:43906014Tue, 06 May 2025 15:09:49 GMT<p>Don&#x27;t be lazy: CompleteP enables compute-efficient deep transformers</p><p></p><p>4 points; 0 discussion replies on HN. <a href="https://hackerpaper.kaipfister.de/p/2505.01618/">Paper and source history</a>. A submission is not an endorsement.</p>cs.LG