mikailkhona
Final year Physics PhD student doing deep learning and computational neuroscience.
MITCambridge, MA
Pinned Repositories
activation_weight_quant
Repo to test various methods and speedups for activation and weight quantization in pytorch
columnformers
A Transformer-inspired model of the brain
CUDA_kernels
Here is a collection of CUDA kernels for small scale simulations. Intended to be used as a guide to learn CUDA programming
dictionary_learning
Double-Descent
Repo to study double descent in linear regression. This is to build intuition.
graph_reps
mikailkhona
Config files for my GitHub profile.
mikailkhona.github.io
Mod_Cog
stepwise_inference_icml24
This repository contains code for ICML 2024: Towards an Understanding of Stepwise Inference in Transformers: A Synthetic Graph Navigation Model
mikailkhona's Repositories
mikailkhona/Mod_Cog
mikailkhona/stepwise_inference_icml24
This repository contains code for ICML 2024: Towards an Understanding of Stepwise Inference in Transformers: A Synthetic Graph Navigation Model
mikailkhona/activation_weight_quant
Repo to test various methods and speedups for activation and weight quantization in pytorch
mikailkhona/columnformers
A Transformer-inspired model of the brain
mikailkhona/CUDA_kernels
Here is a collection of CUDA kernels for small scale simulations. Intended to be used as a guide to learn CUDA programming
mikailkhona/dictionary_learning
mikailkhona/Double-Descent
Repo to study double descent in linear regression. This is to build intuition.
mikailkhona/graph_reps
mikailkhona/mikailkhona
Config files for my GitHub profile.
mikailkhona/mikailkhona.github.io
mikailkhona/Modular_oscillators
mikailkhona/MoE_Interpretability
mikailkhona/muscular_arm
mikailkhona/nanoGPT_extension
mikailkhona/NPEET
Non-parametric Entropy Estimation Toolbox
mikailkhona/pythia
The hub for EleutherAI's work on interpretability and learning dynamics
mikailkhona/Ring_Attractors
mikailkhona/spaghetti
SPAtial GrapHs: nETworks, Topology, & Inference
mikailkhona/sparsegpt
Code for the ICML 2023 paper "SparseGPT: Massive Language Models Can Be Accurately Pruned in One-Shot".
mikailkhona/transformers-qkv-variants