Shreyansh Singh's repositories
Annotated-ML-Papers
Annotations of the interesting ML papers I read
FlashAttention-PyTorch
Implementation of FlashAttention in PyTorch
Speculative-Sampling
Implementation of Speculative Sampling as described in "Accelerating Large Language Model Decoding with Speculative Sampling" by Deepmind
Red-Teaming-Language-Models-with-Language-Models
A re-implementation of the "Red Teaming Language Models with Language Models" paper by Perez et al., 2022
DBMS-Project
DBMS Project (CS361) for Somnath Sevashram
shreyansh26.github.io
My personal website
LLM-Activation-Steering-Experiments
Some experiments with activation steering in LLMs
ML-Paper-Implementations
Implementations of some interesting ML research papers and projects.
RAG-ML-Engg-Open-Book
Query the ML Engineering Open Book using RAG
Tensor-Puzzles-Solutions
Solutions to Tensor puzzles by Sasha Rush - https://github.com/srush/Tensor-Puzzles
VAE-Implementation
A simple implementation of Autoencoder and Variational Autoencoder
Weekend-Projects
Small and interesting projects done in my free time.
CTF-Writeups
CTF (Capture The Flag) writeups, code snippets, notes, scripts
Experiments-with-the-Neural-Tangent-Kernel
Learning more about the NTK through existing paper implementations
Gradient-Descent-on-Neural-Networks-Typically-Occurs-at-the-Edge-of-Stability
A re-implementation of the paper "Gradient Descent on Neural Networks Typically Occurs at the Edge of Stability" by Cohen et al.
hydragen-attention
An implementation of the core attention algorithm in the paper "Hydragen: High-Throughput LLM Inference with Shared Prefixes".
Long-Context-Biencoder
A bi-encoder (sentence/paragraph embedding) model which can work with sequence lengths upto 1024 tokens.
Lottery-Ticket-Hypothesis
A re-implementation of the paper "The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks" by Jonathan Frankle and Michael Carbin
deep_learning_curriculum
Language model alignment-focused deep learning curriculum
LLM-Training-Puzzles-Solutions
The LLM Training Puzzles by Sasha Rush
MLSys-Experiments
A collection of scripts on experimenting and implementing MLSys-related stuff
nanoGPT-kvcache
A fork of nanoGPT which uses KV Cache to do faster inference
resource-stream
CUDA related news and material links
Solving-Substitution-Ciphers-using-MCMC
Solving substitution ciphers using Markov Chain Monte Carlo (MCMC)
Transformer-Puzzles-Solutions
Solutions to the Transformer Puzzles by Sasha Rush - https://github.com/srush/Transformer-Puzzles