Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Aggregate Performance on Average (SST2, AGNEWS, GSM8K, AlpacaEval)
Loading...
35.5
HS Score
SFT
5.756
13.478
21.2
28.922
Jun 10, 2025
HS Score
FA Rate
Updated 1mo ago
Evaluation Results
Method
Method
Links
HS Score
FA Rate
SFT
Backbone=Llama-2-7B
2025.06
35.5
63.1
Lisa-base
Backbone=Llama-2-7B
2025.06
28.8
57.56
BEA
Backbone=Llama-2-7B
2025.06
17.05
62.36
Safe LoRA
Backbone=Llama-2-7B
2025.06
14.5
62.19
SafeInstr
Backbone=Llama-2-7B
2025.06
13.6
59.91
Lisa-aligned
Backbone=Llama-2-7B
2025.06
10.8
63.02
AsFT
Backbone=Llama-2-7B
2025.06
6.9
65.63
Feedback
Search any
task
Search any
task