HackerTrans
TopNewTrendsCommentsPastAskShowJobs

smaddrellmander

90 karmajoined hace 3 meses
idlemachines.co.uk

Submissions

The annotated PyTorch training loop

idlemachines.co.uk
81 points·by smaddrellmander·hace 19 días·19 comments

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

idlemachines.co.uk
1 points·by smaddrellmander·hace 23 días·0 comments

DiffusionGemma: Discrete diffusion in a large language model

idlemachines.co.uk
3 points·by smaddrellmander·el mes pasado·1 comments

Heaven knows I'm perplexed now

idlemachines.co.uk
2 points·by smaddrellmander·el mes pasado·0 comments

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

idlemachines.co.uk
9 points·by smaddrellmander·el mes pasado·0 comments

MAI-Thinking-1: Building a Hill-Climbing Machine [pdf]

microsoft.ai
3 points·by smaddrellmander·el mes pasado·0 comments

Are contrastive losses just cross entropy all along?

idlemachines.co.uk
2 points·by smaddrellmander·el mes pasado·0 comments

Every token, everywhere, all at once

idlemachines.co.uk
2 points·by smaddrellmander·hace 2 meses·0 comments

The cut in the Mixture of Experts compute graph

idlemachines.co.uk
1 points·by smaddrellmander·hace 2 meses·0 comments

DeepSeek V4 from the Inside

idlemachines.co.uk
2 points·by smaddrellmander·hace 2 meses·0 comments

Softmax, can you derive the Jacobian? And should you care?

idlemachines.co.uk
134 points·by smaddrellmander·hace 3 meses·47 comments

Gemma 4 is not your standard transformer

idlemachines.co.uk
2 points·by smaddrellmander·hace 3 meses·0 comments

comments

smaddrellmander
·hace 4 días·discuss
Love this! Cute graphics and the way the balls combine is really fun. Great job.
smaddrellmander
·hace 2 meses·discuss
A similar vein, asking why in PRs. Most people don’t do the above, and asking why when it doesn’t make sense really helps understanding what your colleagues are actually thinking. Helps create a unified perspective on the codebase.