Deterministic, per-token-predictable Chain-of-Thought (CoT) used to supervised-fine-tune a
rank-≤32 LoRA on nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 for the NVIDIA Nemotron Model
Reasoning Challenge (Kaggle).
Authors: Bastien (Tito-42) & Dido Grigorov
Licence: CC BY 4.0 (see NOTICE for upstream terms)
Attribution (required): built on the NVIDIA Nemotron Model Reasoning Challenge; per the competition rules, any use of the competition data must… See the full description on the dataset page:
https://huggingface.co/datasets/Tito-42/nnmrc-cot-datasets.