Simplify

README.md

@@ -18,17 +18,14 @@ tags:
 
 1. Install PyTorch according to your system
 2. `pip install -r requirements.txt`
-3. Retrieval: Cache indices so retrieval will be faster
-3.1 `wget https://huggingface.co/datasets/BeIR/nq/resolve/main/corpus.jsonl.gz`
-3.2 `gunzip corpus.jsonl.gz`
-3.3 Maybe allow creating index here
-4. Clustering: TODO
+3. `wget https://huggingface.co/datasets/BeIR/nq/resolve/main/corpus.jsonl.gz`
+4. `gunzip corpus.jsonl.gz`
+5. Maybe allow creating index here
 
 ## Run
 
 `python app.py`
 
 ## Results
-Results are auto-saved to [mteb/arena-results](https://huggingface.co/datasets/mteb/arena-results).
 
-# python gritlm/rag/eval.py --model_name_or_path GritLM/gritlm-7b --eval_data gritlm/rag/nq_data/test.jsonl --passages corpus.jsonl --save_index_path index_nq
+Results are auto-saved to [mteb/arena-results](https://huggingface.co/datasets/mteb/arena-results).