@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / 0aa38d16f5
Fork 0

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 16, 03:15 UTC

Remove print statement in DeepseekScalingRotaryEmbedding (#11604)

Michael Goin committed Dec 29, 2024 at 15:16 UTC 0aa38d16f56327622c1689d7510171662757deee
1 file changed -1
vllm/model_executor/layers/rotary_embedding.py
-1
@@ -668,7 +668,6 @@ class DeepseekScalingRotaryEmbedding(RotaryEmbedding):
668 cos = (freqs.cos() * self.mscale)
669 sin = (freqs.sin() * self.mscale)
670 cache = torch.cat((cos, sin), dim=-1)
671 - print("Cache shape", cache.shape)
671 return cache
672
673 def forward(