This project implements Low-Rank Adaptation (LoRA) for large language models. The project is structured for modularity and ease of use, allowing for easy downloading, preprocessing, and training of models with various datasets.

1. Clone the repository:

   git clone https://github.com/qlinhta/llm-psl.git

2. Install the required dependencies:

   pip install -r requirements.txt

Model IDs:

Running with Models on E2E Dataset:

python3 ./src/run.py --train_limit [N] --eval_limit [N] --model_id [id] --epochs [ep] --batch_size [b] --learning_rate [lr] --grad_accum_steps [grad] --lora_dim [dim]```