Update README.md

Update README.md

Upload 17 files

remove figures from git tracking

remove figures from git tracking

exclude figures - upload manually to HF

feat: add git LFS for figures

Update Readme

feat: Update Readme

feat: complete 4-method benchmark with honest memory reporting

feat: complete honest 4-method benchmark both models

feat: true Triton 4-bit kernel with real bit packing

chore: Cleanup of the Repo

feat: Mistral results for long context benchmark

feat: Llama results for long context benchmark

fix: Updating the perplexity of my method with FP16

feat: adding visualization for longer context

feat: adding benchmark for longer context

impl: Benchmark results of Llama

impl: Benchmark results of Mistral

feat: Implementation of the vizualize script

feat: Implementing benchmark

feat: auto-save integration results to JSON

feat:Integrating the kernel to the model

feat: Complete working pytorch version of kernel

feat: Testing pure pytorch

chore: clean the quant_cache

chore: libdevice not present in the current version

feat: Initial Triton kernel

feat: sensitivity_map + bit allocation for Llama

Updated bit allocation for mistral

feat: calibration complete + sensitivity heatmap for Mistral-7B

feat: baseline + calibration script