Update answerer.py

answerer.py

@@ -4,21 +4,12 @@ from huggingface_hub import hf_hub_download
 from rwkv.model import RWKV
 from rwkv.utils import PIPELINE, PIPELINE_ARGS
 
-### settings ###
-
-###
-
-
-
-os.environ["RWKV_JIT_ON"] = "1"
-# os.environ["RWKV_CUDA_ON"] = "1" # if "1" then use CUDA kernel for seq mode (much faster)
-
 class Answerer:
   def __init__(self, repo: str, filename: str, vocab: str, strategy: str, ctx_limit: int):
     os.environ["RWKV_JIT_ON"] = "1"
     # os.environ["RWKV_CUDA_ON"] = "1"
     
-    self.__model = RWKV(hf_hub_download(repo, f"{filename}.pth"), strategy=strategy)
+    self.__model = RWKV(hf_hub_download(repo, filename), strategy=strategy)
     self.__pipeline = PIPELINE(self.__model, vocab)
     self.ctx_limit = ctx_limit