Download measured/fin.json from bcv-commons/gram-struct: direct link, hf CLI and curl.
- Browser
- Download file 890 Bytes
-
https://huggingface.co/datasets/bcv-commons/gram-struct/resolve/main/measured/fin.json
- Command line
-
hf download hf://datasets/bcv-commons/gram-struct/measured/fin.json
-
curl -L -o fin.json https://huggingface.co/datasets/bcv-commons/gram-struct/resolve/main/measured/fin.json
890 Bytes
| { | |
| "mechanisms": { | |
| "fertility_priors": { | |
| "date": "2026-09-25", | |
| "enabled": false, | |
| "gold": "helfi/HELFI", | |
| "lambda": 2.0, | |
| "note": "measured 2026-09-25 (roadmap M1), whole Bible vs its own HELFI gold, lambda=2: the intended NEGATIVE CONTROL and it behaved exactly as predicted -- fertility_priors.py flagged 0 anchors (fin has GB074=GB075=1, i.e. BOTH adposition directions attested -> ambiguous -> has_adp=False; GB020-023 all 0 -> has_art=False; GB089=1 -> subject_indexing present -> subj_incomplete=False; every relation gate closed). Result is a pure no-op: F1/exact_span identical to baseline within noise on both conditions (eflomal Delta-0.00065 vs floor 0.0011; +gloss Delta-0.0006 vs floor 0.0008; exact_span moved by -40 and -1.5, i.e. nothing). Confirms the relation-gating logic correctly does nothing when every gate is closed.", | |
| "source": "measured" | |
| } | |
| } | |
| } | |