File size: 1,448 Bytes
9410d8e | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 | CARE-Bench Hugging Face validation report
Status: PASS
Validated files:
- cases_public.jsonl: 439 valid JSONL records
- evaluation_prefixes_public.jsonl: 925 valid JSONL records
- model_inputs_public.jsonl: 925 valid JSONL records
- submission_template_public.csv: 925 data rows
- label_schema.json: valid JSON
- README.md: valid YAML front matter with four viewer configurations
Case splits: {'validation': 93, 'development': 284, 'public_test_1': 62}
Prefix splits: {'validation': 181, 'development': 609, 'public_test_1': 135}
Gold-label counts: {'0A_NO_ESCALATION_INFO_NEEDED': 249, '0B_NO_ESCALATION_SELF_CARE_MONITOR': 162, '1A_ESCALATION_NONURGENT_CARE': 342, '1B_ESCALATION_URGENT_CARE': 172}
Cross-file checks:
- All case_id references resolve to cases_public.jsonl
- Prefix IDs match across evaluation_prefixes, model_inputs, and submission_template
- All primary IDs are unique
SHA-256:
- README.md: c14f5df98f780937a4b2bd5a011b8e8061baf538d34e92403671116b445296ff
- cases_public.jsonl: 510dba567dbc8cb338d91f6c9412ff326be8a6a6bdd1ece241926e5315fe9463
- evaluation_prefixes_public.jsonl: 6d2850f303d05b8f92035c31a11233f41a965e77b3e878e1425f956a00cff534
- label_schema.json: 805c2f21c576c836a081e3636bcee6efceb33a5150c2c2592b9923602a4b75b3
- model_inputs_public.jsonl: 3771e69715278580f9c7d013ac622d333fb0f861c387f2a143517ddb6b60f95b
- submission_template_public.csv: ddd9f65b26a1046a5eb641c0bf16b16deb7c73295a8afe9f1c49bf99c0b3c088
|