Archives | Home

... ...

2026¹

January¹

Cut Cross Entropy: 20x Memory reduction in LLM training through optimized cross entropy kernels

<span title='2026-01-16 08:25:46 +0100 +0100'>January 16, 2026</span> · 8 min · 1566 words · Damilola John

2024³

December²

Understanding Differential Attention.

<span title='2024-12-11 04:14:46 +0100 +0100'>December 11, 2024</span> · 7 min · 1387 words · Damilola John

What do you do when majority of your Coroutines are blocking?

<span title='2024-12-11 04:14:46 +0100 +0100'>December 11, 2024</span> · 1 min · 163 words · Damilola John

June¹

Finetuning GPT2 to Reconstruct Sentences

<span title='2024-06-15 04:14:46 +0100 +0100'>June 15, 2024</span> · 10 min · 2047 words · Damilola John

2023³

August¹

Classifying Code snippets with BERT.

<span title='2023-08-19 04:14:46 +0100 +0100'>August 19, 2023</span> · 4 min · 841 words · Damilola John

July¹

Byte-Pair Encoding, The Tokenization algorithm powering Large Language Models.

<span title='2023-07-20 04:14:46 +0100 +0100'>July 20, 2023</span> · 13 min · 2564 words · Damilola John

May¹

A guide on how AI is changing Computational Photography

<span title='2023-05-10 04:14:46 +0100 +0100'>May 10, 2023</span> · 3 min · 500 words · Damilola John