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

Adding a comment.

Browse Source
This commit is contained in:
Nicolas Patry 2024-01-26 10:30:01 +00:00
parent 97d9ff3a71
commit 7de9141164
1 changed files with 3 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
server/text_generation_server/utils/gptq/exllamav2.py
View File

@ -185,6 +185,9 @@ class QuantLinear(nn.Module):
"g_idx": self.g_idx, "g_idx": self.g_idx,
} }
temp_dq = temp_dq.get_scratch_slice(self.temp_dq_size()) temp_dq = temp_dq.get_scratch_slice(self.temp_dq_size())
# We NEED to keep a pointer on Python side, otherwise the garbage collector will mess with us,
# and `Memory access fault by GPU node-2` will EAT you.
self.temp_dq = temp_dq self.temp_dq = temp_dq
self.q_handle = ext_make_q_matrix(self.q_tensors, temp_dq) self.q_handle = ext_make_q_matrix(self.q_tensors, temp_dq)