Add Typed Decisions benchmark eval results
#1
by SaylorTwift HF Staff - opened
accuracy, kl_from_gold, brier, ece on LocalLLaMA/typed-decisions test split (400 cases, 2000 decisions, one request per case). Independent run 2026-10-16; details: https://huggingface.co/buckets/SaylorTwift/typed-decisions-evals
Add Typed Decisions benchmark eval results
doofz changed pull request status to merged
Hello, we just released https://huggingface.co/HAL-X-AI/THX-02 our second version of model! Would be great for evaluation!