DeepInfra
/

Llama-2-70b-chat-hf-trt-fp8

Model card Files Files and versions Community

yessenzhar commited on Dec 18, 2023

Commit

e4788f3

•

1 Parent(s): dd20dba

bugfix

Files changed (1) hide show

postprocessing/1/model.py +1 -1

postprocessing/1/model.py CHANGED Viewed

@@ -133,7 +133,7 @@ class TritonPythonModel:
             # tokens_batch = tokens_batch.T
             # Postprocessing output data.
-            outputs, output_lens = self._postprocessing(tokens_batch, sequence_lengths)
             # Create output tensors. You need pb_utils.Tensor

             # tokens_batch = tokens_batch.T
             # Postprocessing output data.
+            outputs = self._postprocessing(tokens_batch, sequence_lengths)
             # Create output tensors. You need pb_utils.Tensor