llama.cpp
Fix issue #4791 alloc causes compute_size to be calculated incorrectly in train-text-from-scratch, end result core dump
#5033
Merged
Merged
Fix issue #4791 alloc causes compute_size to be calculated incorrectly in train-text-from-scratch, end result core dump #5033
ggerganov
merged
2 commits into
ggml-org:master
from
bzuzo:bugfix/4791
Fix issue with alloc causing max_compute_size to be calculated
f783c597
bzuzo
changed the title
bzuzo
changed the title 
remove ggml_allocr_free as suggested in issue #4791
38f2e6e7
ggerganov
approved these changes
on 2024-01-19
Login to write a write a comment.
Login via GitHub