423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2025-12-15 03:10:35 +08:00
Code Issues Packages Projects Releases Wiki Activity

resolve gradient checkpointing issue.

Browse Source
This commit is contained in:
Jonery
2024-04-16 12:05:27 +08:00
parent 06c8908d3f
commit 7ecb61822b
4 changed files with 8 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
examples/extras/badam/sft.sh
View File

@@ -31,6 +31,5 @@ python ../../../src/train_bash.py \
--use_badam \
--switch_mode descending \
--badam_verbose 2 \
--switch_block_every 50 \
--pure_bf16 \
--switch_block_every 50