AdaptLLM commited on
Commit
3b82cd3
·
verified ·
1 Parent(s): ec65cdc

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -2
README.md CHANGED
@@ -156,7 +156,7 @@ You can use the following scripts to reproduce our results and evaluate any othe
156
  DOMAIN='biomedicine'
157
 
158
  # Specify any Huggingface model name (Not applicable to chat models)
159
- MODEL='AdaptLLM/medicine-LLM'
160
 
161
  # Model parallelization:
162
  # - Set MODEL_PARALLEL=False if the model fits on a single GPU.
@@ -171,7 +171,7 @@ You can use the following scripts to reproduce our results and evaluate any othe
171
  # - Set to False for AdaptLLM.
172
  # - Set to True for instruction-pretrain models.
173
  # If unsure, we recommend setting it to False, as this is suitable for most LMs.
174
- add_bos_token=False
175
 
176
  # Run the evaluation script
177
  bash scripts/inference.sh ${DOMAIN} ${MODEL} ${add_bos_token} ${MODEL_PARALLEL} ${N_GPU}
 
156
  DOMAIN='biomedicine'
157
 
158
  # Specify any Huggingface model name (Not applicable to chat models)
159
+ MODEL='instruction-pretrain/medicine-Llama3-8B'
160
 
161
  # Model parallelization:
162
  # - Set MODEL_PARALLEL=False if the model fits on a single GPU.
 
171
  # - Set to False for AdaptLLM.
172
  # - Set to True for instruction-pretrain models.
173
  # If unsure, we recommend setting it to False, as this is suitable for most LMs.
174
+ add_bos_token=True
175
 
176
  # Run the evaluation script
177
  bash scripts/inference.sh ${DOMAIN} ${MODEL} ${add_bos_token} ${MODEL_PARALLEL} ${N_GPU}