Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Fine-tuning on OpenHermes
Loading...
0.706
Evaluation Loss
RefLoRA
0.70424
0.71612
0.728
0.73988
May 29, 2026
Evaluation Loss
Updated 2d ago
Evaluation Results
Method
Method
Links
Evaluation Loss
RefLoRA
Backbone=Qwen-2.5-3B,...
2026.05
0.706
LoRA
Backbone=Qwen-2.5-3B,...
2026.05
0.707
DoRA
Backbone=Qwen-2.5-3B,...
2026.05
0.707
LoRA-RITE
Backbone=Qwen-2.5-3B,...
2026.05
0.707
BaLoRA
Backbone=Qwen-2.5-3B,...
2026.05
0.707
OLoRA
Backbone=Qwen-2.5-3B,...
2026.05
0.712
LoRO
Backbone=Qwen-2.5-3B,...
2026.05
0.75
Feedback
Search any
task
Search any
task
