transformations-english-danish

Chat-template-ready synthetic instruction-transformation examples generated with google/gemma-4-31B-it. Each example transforms English source text into a Danish response.

Contents

• Format: gzip-compressed JSON Lines under data/*.jsonl.gz
• Schema: {"messages": [{"role": "user", "content": "..."}, {"role": "assistant", "content": "..."}]}
• Files: 409
• Rows: 246,288
• Language pair: en_da

Actual Sources

The exported rows contain only chat messages; aggregate provenance below was reconstructed by exact matching against accepted generation metadata.

Source cluster	Rows
DFM4 scientific/arXiv summarization seeds	237,965
DBC converted sources	7,671
facebook/asset	369
LexDK converted sources	283

Task Families

Task family	Rows
non_copy_rewrite	49,983
past_tense_rewrite	49,887
exact_sentence_summary	49,383
numbered_fact_extraction	49,351
child_friendly_simplification	47,684

Export Checks

Check	Rows
Exported rows	246,288
Matched to generation metadata	246,288
Unmatched exported rows	0
Accepted regeneration rows in export	1,260

Example

{
  "messages": [
    {
      "role": "user",
      "content": "Opgave: Forenkl passagen for et barn. Bevar de vigtigste fakta. Svarsprog: dansk. Strengt format: Returnér kun den forenklede tekst. Skriv ikke formuleringer som 'Her er' eller 'Det betyder'. Input: The article reports on the computer accessories available for the second-half back-to-school and holiday shopping seasons. A flush market for add-on items is expected to continue with computer sales expected to increase 8.5 percent in 2006. The available computer accessories include a new line of notebook computer locks from Kensington, Iogear' wireless multimedia keyboard/mouse combo, Belkin's N1 wireless router and the Chillhub notebook cooling pad from Targus."
    },
    {
      "role": "assistant",
      "content": "Der kommer mange nye ting til computere, som man kan købe, når skolen starter, og når det bliver jul. Flere mennesker forventes at købe computere i 2006. Man kan blandt andet købe låse til bærbare computere, trådløse tastaturer og mus, trådløse routere og køleplader til bærbare computere."
    }
  ]
}

Filtering

Rows were generated and judged with google/gemma-4-31B-it. Rejected or unjudged rows were excluded; accepted regeneration rows replace their original generated rows.

Exact byte-for-byte recreation is not expected because teacher sampling, serving order, and judge decisions can vary. Seed selection, task order, and prompt templates are deterministic for a fixed seed.

Machine-readable details are in generation_config.json, accepted_selection_summary.json, and audit_summary.json.

Recreate

Start an OpenAI-compatible teacher endpoint, then run a small smoke sample:

python recreate_dataset.py \
  --base-url http://127.0.0.1:8100/v1 \
  --model posttrain-gemma-teacher \
  --rows 1000 \
  --output generated/train.jsonl.gz

By default, the script uses seeds/source_texts.jsonl.gz and writes only judge-accepted rows.