Datasetpublic

japhba/loracles-fineweb-multidoc-qa-1704

loracles-finetune-gemini-3.1-flash-lite Synthetic Loracle supervision data generated from FineWeb. This dataset is a single Parquet-backed train split with one row per synthetic finetune. This upload is a partial snapshot of a larger run. Run summary source dataset: HuggingFaceFW/fineweb / sample-10BT / train sampled docs: 15000 synthetic finetunes: 3009 generated finetunes uploaded: 2018 generator backend: openrouter generator model:… See the full description on the dataset page: https://huggingface.co/datasets/japhba/loracles-fineweb-multidoc-qa-1704.

sourceHugging Faceupdated 5mo agoView on Hugging Face

0likes39downloads

Dataset Card

loracles-finetune-gemini-3.1-flash-lite

Synthetic Loracle supervision data generated from FineWeb.

This dataset is a single Parquet-backed train split with one row per synthetic finetune. This upload is a partial snapshot of a larger run.

Run summary

—source dataset: HuggingFaceFW/fineweb / sample-10BT / train
—sampled docs: 15000
—synthetic finetunes: 3009
—generated finetunes uploaded: 2018
—generator backend: openrouter
—generator model: google/gemini-3.1-flash-lite-preview
—max docs per finetune: 20
—max token budget per finetune: 10000
—questions per finetune: 10

Table schema

—one row per synthetic finetune
—nested questions field with 10 ordered question-answer pairs
—finetune metadata merged in from synthetic_models.jsonl
—stored on the Hub as Parquet

Local artifacts

—run directory: /ceph/scratch/jbauer/loracle/fineweb_openrouter_flash31lite_large_20260417
—sample manifest: sample_manifest.json
—question-group shards uploaded: 3
—generation manifests uploaded: 2
—auxiliary files kept as repo artifacts: synthetic_models.jsonl, sampled_docs.jsonl, question_failures.shard-*.jsonl