List your company

Safety Alignment on WildJailbreak (Safe@1)

77.2Safe@1R1 - 8B + UnsafeChain full
33.5244.8656.267.54Jul 29, 2025Sep 8, 2025Oct 20, 2025Dec 1, 2025Jan 11, 2026Feb 22, 2026Apr 5, 2026

Evaluation results

MethodLinks
R1 - 8B + UnsafeChain fullSize=8B, Alignment=UnsafeChain (full)2025.0777.2
R1 - 7B + UnsafeChain fullSize=7B, Alignment=UnsafeChain (full)2025.0774.8
R1 - 8B + UnsafeChain selectedSize=8B, Alignment=UnsafeChain (selected subset)2025.0773.6
R1 - 8B + SafeChainSize=8B, Alignment=SafeChain2025.0768
R1 - 8B + STAR-1Size=8B, Alignment=STAR-12025.0767.6
R1 - 8B + UnsafeChain randomSize=8B, Alignment=UnsafeChain (random subset)2025.0765.6
R1 - 7B + STAR-1Size=7B, Alignment=STAR-12025.0765.2
R1 - 8BSize=8B2025.0764.8
R1 - 1.5B + UnsafeChain fullSize=1.5B, Alignment=UnsafeChain (full)2025.0764.8
R1 - 7B + SafeChainSize=7B, Alignment=SafeChain2025.0762.4
R1 - 7B + UnsafeChain selectedSize=7B, Alignment=UnsafeChain (selected subset)2025.0762
R1 - 7B + UnsafeChain randomSize=7B, Alignment=UnsafeChain (random subset)2025.0761.6
R1 - 7BSize=7B2025.0760.8
SIFTRuntime (min)=215.62026.0456
BLURRuntime (min)=114.72026.0454.4
AdamWRuntime (min)=110.32026.0452.8
MuonRuntime (min)=191.22026.0452
POMERuntime (min)=115.42026.0451.6
R1 - 1.5B + SafeChainSize=1.5B, Alignment=SafeChain2025.0750.8
Base modelRuntime (min)=N/A2026.0447.6
R1 - 1.5BSize=1.5B2025.0738.4
R1 - 1.5B + UnsafeChain randomSize=1.5B, Alignment=UnsafeChain (random subset)2025.0737.2
R1 - 1.5B + STAR-1Size=1.5B, Alignment=STAR-12025.0736.4
R1 - 1.5B + UnsafeChain selectedSize=1.5B, Alignment=UnsafeChain (selected subset)2025.0735.2