1 results

chain-determinism

inall models datasets apps agents people projects

Datasets

All datasets matching “chain-determinism”

cruxia /chain-determinism-bench-v1 Chain-Determinism Bench v1 Cross-vendor measurements of LLM agent-loop chain-determinism: 31,764 hash-only observations spanning 11 vendors across four study phases. Released by Cruxia-AI under CC-BY-4.0. What this is 31,764 hash-only observations from a 22,000-run multi-vendor chain-determinism study plus follow-up phases (Phase 1b temperature sweep, Phase 2 mechanism attribution, Phase 3.5 SWE-bench Lite). Each row records what an agent loop produced for one (vendor… See the full description on the dataset page: https://huggingface.co/datasets/cruxia/chain-determinism-bench-v1.tabularother10K<n<100K1 likes14 downloads4mo agoHugging Face