Unclaimed profileBuilt from public GitHub and Hugging Face activity. Shreyansh has not joined Lounge yet.Claim it
Not on Lounge yet
Shreyansh Singh
developerInference-maxxing and agent harnesses at Level AI. Previously @Mastercard AI Garage, @Samsung Research.
CS, IIT (BHU) Varanasi.
Public projects
Annotated-ML-Papers
Annotations of the interesting ML papers I read
annotated-paperbertdeep-learning
ML-Optimizers-JAX
Toy implementations of some popular ML optimizers using Python/JAX
Pythonadamadam-optimizergradient-descent
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
Pythonadversarial-attacksadversarial-machine-learninggpt-2
LLM-Sampling
A collection of various LLM sampling methods implemented in pure Pytorch
Pythonllmllm-inferencesampling-methods
An-Empirical-Model-of-Large-Batch-Training
An approximate implementation of the OpenAI paper - An Empirical Model of Large-Batch Training for MNIST
Python
STILL-Towards-Infinite-Context-Windows
An implementation of the Baseten blog - Towards infinite context windows: neural KV cache compaction
Python
