A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 17, 06:45 UTC
Fix typo (#1383)
Wang Ran (汪然) committed
Oct 17, 2023 at 12:53 UTC
a132435204aac8506e41813f90d08ddf7eca43b2
1 file changed
+3
-3
vllm/model_executor/input_metadata.py
+3
-3
@@ -79,7 +79,7 @@ class InputMetadata:
79
f'prompt_lens={self.prompt_lens}, '
80
f'num_generation_tokens={self.num_generation_tokens}, '
81
f'context_lens={self.context_lens}, '
82
- f'max_context_len={self.max_context_len}), '
82
+ f'max_context_len={self.max_context_len}, '
83
f'max_num_blocks_per_seq={self.max_num_blocks_per_seq}, '
84
- f'block_tables={self.block_tables}), '
85
- f'slot_mapping={self.slot_mapping}')
84
+ f'block_tables={self.block_tables}, '
85
+ f'slot_mapping={self.slot_mapping})')