# 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):** 19 (median 19.0, min 15, max 23)
- **Schema errors:** 0 (should be 0)
- **File size:** 0.00 MB
- **SHA256 (data.jsonl):** `a5b5f45032610f3754654b9f5c688c1e9fc7926a93c8aa2f5e4f0fb34e35eedc`

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