neulab
/

t-dig

@@ -41,7 +41,7 @@ These are the recommended checkpoints for use, and differ in places from the pap
 ## Use
-The policy expects the instruction the paper trained it on; `tdig.instruction` in the [code repository](https://github.com/neulab/t-dig) builds it. Generation, with vLLM:
 ```python
 from vllm import LLM, SamplingParams

 ## Use
+The policy expects the instruction the paper trained it on; `tdig.instruction` in the [code repository](https://github.com/simran-khanuja/t-dig) builds it. Generation, with vLLM:
 ```python
 from vllm import LLM, SamplingParams