Update README.md

README.md

@@ -88,7 +88,7 @@ You can use the following scripts to reproduce our results and evaluate any othe
    DOMAIN='finance'
   
    # Specify any Huggingface model name (Not applicable to chat models)
-   MODEL='AdaptLLM/finance-LLM'
+   MODEL='instruction-pretrain/finance-Llama3-8B'
   
    # Model parallelization:
    # - Set MODEL_PARALLEL=False if the model fits on a single GPU. 
@@ -103,7 +103,7 @@ You can use the following scripts to reproduce our results and evaluate any othe
    # - Set to False for AdaptLLM.
    # - Set to True for instruction-pretrain models.
    # If unsure, we recommend setting it to False, as this is suitable for most LMs.
-   add_bos_token=False
+   add_bos_token=True
 
    # Run the evaluation script
    bash scripts/inference.sh ${DOMAIN} ${MODEL} ${add_bos_token} ${MODEL_PARALLEL} ${N_GPU}