smaddrellmander | Dark Hacker News

user:	smaddrellmander
created:	April 21, 2026
karma:	97
about:	idlemachines.co.uk

1.	Fast weights and sparse attention in GLM-5.3-Flash(idlemachines.co.uk) 7 points by smaddrellmander 5 days ago \| 0 comments
2.	Tokenisation: Who decides what a token is anyway?(idlemachines.co.uk) 2 points by smaddrellmander 29 days ago \| 0 comments
3.	Adam and AdamW: adaptive optimisation and weight decay(idlemachines.co.uk) 1 points by smaddrellmander 39 days ago \| 0 comments
4.	Thinking Machines dropped RoPE, and it's a good idea(idlemachines.co.uk) 6 points by smaddrellmander 48 days ago \| 0 comments
5.	Nesso-1: Accelerating Open-Source Binding Affinity Predictions [pdf](valencelabs.com) 2 points by smaddrellmander 50 days ago \| 1 comment
6.	The annotated PyTorch training loop(idlemachines.co.uk) 81 points by smaddrellmander 78 days ago \| 19 comments
7.	Mae vs. MSE: more than just the mean vs. median debate(idlemachines.co.uk) 1 points by smaddrellmander 82 days ago \| 0 comments
8.	DiffusionGemma: Discrete diffusion in a large language model(idlemachines.co.uk) 3 points by smaddrellmander 89 days ago \| 1 comment
9.	Heaven knows I'm perplexed now(idlemachines.co.uk) 2 points by smaddrellmander 94 days ago \| 0 comments
10.	Reading MAI's efficiency gain. How to pick architectures like serious people(idlemachines.co.uk) 9 points by smaddrellmander 97 days ago \| 0 comments
11.	MAI-Thinking-1: Building a Hill-Climbing Machine [pdf](microsoft.ai) 3 points by smaddrellmander 98 days ago \| 0 comments
12.	Are contrastive losses just cross entropy all along?(idlemachines.co.uk) 2 points by smaddrellmander 99 days ago \| 0 comments
13.	Every token, everywhere, all at once(idlemachines.co.uk) 2 points by smaddrellmander 106 days ago \| 0 comments
14.	The cut in the Mixture of Experts compute graph(idlemachines.co.uk) 1 points by smaddrellmander 113 days ago \| 0 comments
15.	DeepSeek V4 from the Inside(idlemachines.co.uk) 2 points by smaddrellmander 120 days ago \| 0 comments
16.	Softmax, can you derive the Jacobian? And should you care?(idlemachines.co.uk) 134 points by smaddrellmander 134 days ago \| 47 comments
17.	Gemma 4 is not your standard transformer(idlemachines.co.uk) 2 points by smaddrellmander 140 days ago \| 0 comments