Updated execution guide

Signed-off-by: Jonathan Bnayahu <[email protected]>

src/about.py

@@ -89,7 +89,8 @@ To reproduce our results, here is the commands you can run:
 
 ```
 pip install unitxt[bluebench]
-unitxt-evaluate --tasks "benchmarks.bluebench" --model cross_provider --model_args "model_name=$MODEL_TO_EVALUATE_IN_LITELLM_FORMAT,max_tokens=256" --output_path ./results/bluebench --log_samples --trust_remote_code --batch_size 8
+unitxt-evaluate --tasks "benchmarks.bluebench" --model cross_provider --model_args "model_name=MODEL_TO_EVALUATE_IN_LITELLM_FORMAT,max_tokens=1024" --output_path ./results/bluebench --log_samples --trust_remote_code --batch_size 8
+unitxt-summarize ./results/bluebench
 ```
 """