Peter Kim

Expert
@tspeterkim

I like being useful.

tspeterkim.github.io/

flash-attention-minimal. Flash Attention in ~100 lines of CUDA (forward pass only)

1.2k

insta-chat. DIY Instagram Chat Automation with Google Sheets

239

paged-attention-minimal. a minimal cache manager for PagedAttention, on top of llama3.

149

mixed-precision-from-scratch. Mixed precision training from scratch with Tensors and CUDA

30

cuda-1brc. My CUDA solution to the 1BRC

11

attentive-reader. TensorFlow-based implementation of Stanford's Attentive Reader for the CNN/DailyMail datatset

8

lucene-wikipedia. Lucene-based Multi-sentence context retriever for open-domain QA setting

4

diffusion. the new transformers for generative models?

2

Spoon-Knife. This repo is for demonstration purposes only.

2

tweet-emoji-predictor. Tweet Emoji Predictor - SemEval 2018 Task 2

2

style-transferer. Lightweight style transferer built with SqueezeNet

1

Spread-The-Word. The source code for Spread The Word.

1

autograd. autograd engine with basic neural net apis

1

Hello-World. My First Repo

1

CS118-Lab2. C++

1

positioning. RNN based prediction of wearer's position given sensor data from Android Wear

1

Free-For-Sale. PHP

1

cuda-matmult. Cuda

1

ml-algorithms. from the scratch

1

gpt. Jupyter Notebook

1
20
Apply