structured-outputs-calibration-v1

> [!TIP] > Support this work: **[donate.sybilsolutions.ai](https://donate.sybilsolutions.ai)** > > REAP surfaces: [GLM](https://huggingface.co/spaces/0xSero/reap-glm-family) | [MiniMax](https://huggingface.co/spaces/0xSero/reap-minimax-family) | [Qwen](https://huggingface.co/spaces/0xSero/reap-qwen-family) | [Gemma](https://huggingface.co/spaces/0xSero/reap-gemma-family) | [Paper](https://arxiv.org/abs/2510.13999) | [Code](https://github.com/CerebrasResearch/reap) | [PR17](https://github.com/CerebrasResearch/reap/pull/17) | [Cerebras Collection](https://huggingface.co/collections/cerebras/cerebras-reap) # structured-outputs-calibration-v1 Structured-output calibration set for REAP observer runs, focused on preserving: - strict JSON generation - schema-conditioned JSON responses - Mermaid diagram generation - fenced Mermaid block formatting ## Contents - `data.jsonl`: normalized calibration records - `summary.json`: build summary with source counts - `config.yaml`: source mix used to produce the dataset ## Source mix - `nvidia/Nemotron-RL-instruction_following-structured_outputs` — 160 - `Arun63/sharegpt-structured-output-json` — 30 - `Celiadraw/text-to-mermaid` — 120 - `sts07142/mermaid_samples_13k` (`wrapped`) — 120 Total samples: `430` ## Record format Each JSONL row contains: - `id` - `source_id` - `repo_id` - `subset` - `split` - `text` `text` is normalized into prompt/response-style plain text suitable for calibration / observation workloads.