EPFL-LARA/lean-world-model-sft
Supervised fine-tuning corpus The feedback-annotated SFT collection for Formal Theorem Proving with Lean World Models, containing direct proofs and proof-repair histories. Split Rows Contents train 2,763,895 Supervised prompt/answer pairs test 623,741 Unsolved-problem source catalog Load from datasets import load_dataset examples = load_dataset("EPFL-LARA/lean-world-model-sft", split="train", streaming=True) row = next(iter(examples)) prompt… See the full description on the dataset page: https://huggingface.co/datasets/EPFL-LARA/lean-world-model-sft.
| Author / organization | EPFL-LARA |
|---|---|
| Downloads | 0 (as reported by Hugging Face) |
| Likes | 0 |
| Created | 16 October 2026 |
| Last updated at source | 16 October 2026 |
Popularity, last 2 days
0 now. Counts as reported by the source, snapshotted daily by Silytics.
Connected in the knowledge graph
- EPFL-LARA/lean-standard-prover-qwen3-8b-grpoModel · used to train
- EPFL-LARA/lean-world-model-qwen3-4b-grpoModel · used to train
- mradermacher/lean-world-model-qwen3-8b-reasoning-grpo-GGUFModel · used to train
- EPFL-LARA/lean-standard-prover-qwen3-8b-reasoning-grpoModel · used to train
- mradermacher/lean-standard-prover-qwen3-8b-reasoning-grpo-GGUFModel · used to train
- EPFL-LARA/lean-world-model-qwen3-8b-reasoning-grpoModel · used to train
- EPFL-LARA/lean-world-model-qwen3-8b-sftModel · used to train
- EPFL-LARA/lean-world-model-qwen3-8b-grpoModel · used to train
- EPFL-LARA/lean-standard-prover-qwen3-8b-sftModel · used to train
More from EPFL-LARA
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.