# Subnet 96 — Clean Q/A Dataset

**Format:** one JSONL per line:
```json
{"system": null, "conversations":[{"role":"user","content":"..."}, {"role":"assistant","content":"..."}]}
```

- **Total pairs:** 2
- **Avg answer length (tokens):** 30 (median 30.0, min 26, max 34)
- **Schema errors:** 0 (should be 0)
- **File size:** 0.00 MB
- **SHA256 (data.jsonl):** `25a907653bebccf712108c7c6a6c06499653e2d0f6e8237b3a374ad2933620a7`

> Language: **English**
> Intended for: **Bittensor Subnet 96 validators**
> Generation: local LLaMA (GPU) + cleaning/normalization scripts.