periodic: reasoning train.enhancer_editing.reasoning.jsonl

T3 oracle-filtered top-1 dataset (random-edit + oracle-rank for activity_boost/cell_type_transfer; gold kept for promoter_retarget)

v15 unified+mdlm+aux T1 final ckpt: final/tokenizer.json

Upload motif_targets/from_tools_tier1/extracted.jsonl with huggingface_hub

Upload motif_targets/from_tools_uncapped/extracted.jsonl with huggingface_hub

Upload motif_targets/merged_full/extracted.jsonl with huggingface_hub

periodic: reasoning train.pair_prediction.reasoning.frozen_20260428T112500Z.jsonl

periodic: reasoning train.enhancer_generation.reasoning.frozen_20260428T112500Z.jsonl

reasoning train.enhancer_generation.reasoning.jsonl 5768 rows (cycle 1)

migration: tokenizer.json

reasoning: train.pair_prediction.reasoning.jsonl (live snapshot)

prod_samples: test.pair_prediction.strat7c.n7k.jsonl

prod_samples: test.enhancer_editing.strat7c.n7k.jsonl

prod_samples: test.enhancer_generation.strat7c.n7k.jsonl

prod_samples: train.joint_multitask_balanced.strat7c.n105k.jsonl

prod_samples: train.joint_multitask.strat7c.jsonl

upload: T2 full test enriched (6.7GB)

prod_samples: train.pair_prediction.strat7c.n200k.jsonl

prod_samples: train.pair_prediction.strat7c.n35k.jsonl

prod_samples: train.enhancer_editing.strat7c.n35k.jsonl

prod_samples: train.enhancer_generation.strat7c.n35k.jsonl

upload: T3 full test enriched (3.7GB)

upload: T1 full test enriched (3.5GB)

Upload data/t2_regen_enhancer_scan/jsonl/train.pair_prediction.jsonl with huggingface_hub

Upload data/t2_regen_enhancer_scan/jsonl/test.pair_prediction.jsonl with huggingface_hub

Upload data/full_enriched_v2_with_enh_scan/jsonl/test.pair_prediction.jsonl with huggingface_hub

initial commit