This website requires JavaScript.
Explore
Help
Sign In
cmedina
/
agenx-lora-training
Watch
1
Star
0
Fork
0
You've already forked agenx-lora-training
Code
Issues
Pull Requests
Actions
Packages
Projects
Releases
Wiki
Activity
Files
0456d55c8adf5ed72b589330e2d9ecf73561fb2b
agenx-lora-training
/
training
/
configs
History
Christian Medina
0456d55c8a
fix: disable gradient checkpointing (device mismatch with distributed MoE)
2026-07-02 17:29:19 -04:00
..
ds_zero3.json
feat: use DeepSpeed ZeRO-3 for proper model sharding across GPUs
2026-07-01 22:08:37 -04:00
ornith-35b-lora.yaml
fix: disable gradient checkpointing (device mismatch with distributed MoE)
2026-07-02 17:29:19 -04:00