Jianshu001/arabic-daily-batch01-smoke20-rewriter
Batch 01 — Smoke 20 (Gemma-as-rewriter) 20-record smoke validating the Gemma-as-rewriter cleanup architecture. Approach Any prompt given to Gemma gets echoed into its own thinking field. Instead of fighting the echo, we isolate it by using Gemma as a rewriter: Cascade-regenerate assistant turn → (thinking1, answer) Send thinking1 to Gemma with a cleanup instruction → (thinking2, answer2) Discard thinking2 (absorbs the cleanup-instruction echo) Keep answer2 as the… See the full description on the dataset page: https://huggingface.co/datasets/Jianshu001/arabic-daily-batch01-smoke20-rewriter.
Batch 01 — Smoke 20 (Gemma-as-rewriter)
20-record smoke validating the Gemma-as-rewriter cleanup architecture.
Approach
Any prompt given to Gemma gets echoed into its own thinking field. Instead of fighting the echo, we isolate it by using Gemma as a rewriter:
- Cascade-regenerate assistant turn → (thinking1, answer)
- Send thinking1 to Gemma with a cleanup instruction → (thinking2, answer2)
- Discard thinking2 (absorbs the cleanup-instruction echo)
- Keep answer2 as the cleaned thinking
Same pattern applied to assistant text when AI self-reference is detected.
Results
Schema
- User message fields: turn, role, text
- Assistant message fields: turn, role, thinking, text
No audit fields included — textoriginal, thinkingoriginal, rewritelog, and thinkingoriginallang are all stripped before upload.