data/processed/preprocess_summary.json

{
  "counts": {
    "evidence_corpus.jsonl": 667,
    "retrieval_pairs_train.jsonl": 2537,
    "retrieval_pairs_validation.jsonl": 939,
    "verifier_test.jsonl": 169,
    "verifier_train.jsonl": 916,
    "verifier_validation.jsonl": 168
  },
  "retrieval_sources": [
    "andreiaalexa/scifact-relevance-pairs"
  ],
  "sources": [
    "allenai/scifact_entailment"
  ]
}