423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2026-04-21 20:36:02 +08:00
Code Issues Packages Projects Releases Wiki Activity

[v1] add deepspeed zero3 trigger for low memory usage weight loading (#10300)

Browse Source
This commit is contained in:
jiaqiw09
2026-04-21 14:09:52 +08:00
committed by GitHub
parent f5d739b132
commit 28a6ea1cdc
13 changed files with 160 additions and 18 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
examples/v1/train_full/train_full_fsdp2.yaml
View File

@@ -21,7 +21,6 @@ output_dir: outputs/test_fsdp2
micro_batch_size: 1
cutoff_len: 2048
learning_rate: 1.0e-4
bf16: false
max_steps: 10
### sample