This dataset is a 979-row instruction distillation dataset generated using NVIDIA's flagship Nemotron-3-Ultra-550B model. The prompt source is a randomized subset of the high-quality ianncity/General-Distillation-Prompts-1M corpus.

Dataset Details

Size: 979 rows Columns: index, instruction, output Generator Model: nvidia/nemotron-3-ultra-550b-a55b Formatting: Output formats preserve reasoning paths wrapped inside… See the full description on the dataset page: https://huggingface.co/datasets/SLoonker/Nemotron-Ultra-1000x.