huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-11 04:14:52 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix wrong refactor of rope

Browse Source
This commit is contained in:
Felix Marty 2023-11-14 10:43:02 +00:00
parent 8617d4795a
commit a992084b9b
1 changed files with 11 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

23
server/text_generation_server/utils/layers.py
View File

@ -595,21 +595,20 @@ try:
elif IS_ROCM_SYSTEM:
# For RoCm, we fall back on a manual implementation given that Flash Attention's ROPE kernel can not be compiled for RoCm.
# We could use VLLM ROPE kernel here (compatible with RoCm), but the API is different and would require position_ids: https://github.com/vllm-project/vllm/blob/1a2bbc930135cd3b94fbff2aafbdf5c568acc8bd/csrc/pos_encoding.cpp#L3
def rope_forward_rocm(self, x: torch.Tensor, cos: torch.Tensor, sin: torch.Tensor):
rotary_dim = cos.shape[-1]
rotary_dim = cos.shape[-1]
dtype = x.dtype
x_upcast = x.to(torch.float32)
cos = cos.to(torch.float32)
sin = sin.to(torch.float32)
dtype = x.dtype
x_upcast = x.to(torch.float32)
cos = cos.to(torch.float32)
sin = sin.to(torch.float32)
x1 = x_upcast[..., :rotary_dim]
x2 = x_upcast[..., rotary_dim : 2 * rotary_dim]
x1 = x_upcast[..., :rotary_dim]
x2 = x_upcast[..., rotary_dim : 2 * rotary_dim]
# Flash Attention rotary_emb kernel casts everything to float, not sure why, so we do so here as well.
x[..., :rotary_dim] = (x1 * cos - x2 * sin).to(dtype)
x[..., rotary_dim : 2 * rotary_dim] = (x1 * sin + x2 * cos).to(dtype)
return x
# Flash Attention rotary_emb kernel casts everything to float, not sure why, so we do so here as well.
x[..., :rotary_dim] = (x1 * cos - x2 * sin).to(dtype)
x[..., rotary_dim : 2 * rotary_dim] = (x1 * sin + x2 * cos).to(dtype)
return x
else:
raise ValueError("Your system seem to be not supported. Please check your install or open an issue at https://github.com/huggingface/text-generation-inference/issues with a clear reproduction.")