smaddrellmander

Joined 6 months ago

Karma

99

Submissions

23

Member since

Apr 21

idlemachines.co.uk

View on Hacker News

Recent submissions

Efficient Streaming Language Models with Attention Sinks idlemachines.co.uk

2 smaddrellmander 3 days ago 0 comments

Learning Without Backpopagation idlemachines.co.uk

6 smaddrellmander 12 days ago 1 comments

How to quickly get prepared for your technical round idlemachines.co.uk

1 smaddrellmander 20 days ago 1 comments

Fast weights and sparse attention in GLM-5.3-Flash idlemachines.co.uk

7 smaddrellmander 1 month ago 0 comments

Tokenisation: Who decides what a token is anyway? idlemachines.co.uk

2 smaddrellmander 2 months ago 0 comments

Adam and AdamW: adaptive optimisation and weight decay idlemachines.co.uk

1 smaddrellmander 3 months ago 0 comments

Thinking Machines dropped RoPE, and it's a good idea idlemachines.co.uk

6 smaddrellmander 3 months ago 0 comments

Comment · 3 months ago

Curious what people think? Decent binding affinity model out of Valence Labs.

Nesso-1: Accelerating Open-Source Binding Affinity Predictions [pdf] valencelabs.com

2 smaddrellmander 3 months ago 1 comments

Comment · 3 months ago

Love this! Cute graphics and the way the balls combine is really fun. Great job.

The annotated PyTorch training loop idlemachines.co.uk

81 smaddrellmander 4 months ago 19 comments

Mae vs. MSE: more than just the mean vs. median debate idlemachines.co.uk

1 smaddrellmander 4 months ago 0 comments

DiffusionGemma: Discrete diffusion in a large language model idlemachines.co.uk

3 smaddrellmander 4 months ago 1 comments

Heaven knows I'm perplexed now idlemachines.co.uk

2 smaddrellmander 4 months ago 0 comments

Reading MAI's efficiency gain. How to pick architectures like serious people idlemachines.co.uk

9 smaddrellmander 5 months ago 0 comments