Intrinsically aligned models via CL-BPL. Ethics in weights, not system prompts. 4B ranks 25th globally for IF on LiveBench.
Lethean
lthn
AI & ML interests
None yet
Recent Activity
updated a dataset about 5 hours ago
lthn/LEM-benchmarks updated a model about 5 hours ago
lthn/lemer updated a model about 8 hours ago
LetheanNetwork/lemer-pt