Update README.md with the link direct to the technical report

Update README.md with the citation section

trainable hyperparamger (lora rank, adapter size) plot script

rename experiment type for model size from type 3

add support to analyze the impact of data size on LoRA, adapter and fine-tuning

update README about HPC notes

update README about HPC notes

update README.md

refactor hp search expr by adding sorting data frame by rows and columns values and also output latex

fixed fine tuning bug that hangs due to internal accelerate issues by calling on all processes

add scripts forr finetuning grid hp serach

backward compatibility for restarting expr: zero-init optimization step, end training based on global step

bash scrips changes for lora and prefix tuning

fixed adapter scripts for easier rerun

Validation ni adapterpeft cluster customized training (#4)