Upload README.md with huggingface_hub

README.md

@@ -23,19 +23,19 @@ was measured by the run that produced the data.
 
 ## What is in here
 
-- **1636** generated learner sentences (`raw/raw_texts.parquet`)
-- **1578** accepted gradings (`raw/raw_labels.parquet`)
-- **190** distinct dictionary senses
+- **12454** generated learner sentences (`raw/raw_texts.parquet`)
+- **12103** accepted gradings (`raw/raw_labels.parquet`)
+- **2000** distinct dictionary senses
 
 ### Splits
 
 | Split | Rows |
 |---|---:|
-| `train` | 1225 |
-| `val` | 179 |
-| `test` | 174 |
+| `train` | 9545 |
+| `val` | 1251 |
+| `test` | 1307 |
 
-Grouped by target word, not by row. Sentences appearing in more than one split: **0**. Rejected during validation: **0** of 1578.
+Grouped by target word, not by row. Sentences appearing in more than one split: **0**. Rejected during validation: **0** of 12103.
 
 ```json
 {
@@ -73,12 +73,12 @@ a correct row and a wrong one look identical.
 
 | | |
 |---|---|
-| Teacher model | `claude-opus-5` |
+| Teacher model | `zm1/zyloo/gemini-3.1-pro-preview` |
 | Endpoint | `https://api.vilao.ai/v1` |
-| Call 1 requests | 190 |
-| Call 2 requests | 190 |
-| Format validity | 0.6947 |
-| Batch diversity (distinct-2) | 0.9555 |
+| Call 1 requests | 6 |
+| Call 2 requests | 6118 |
+| Format validity | 0.9475 |
+| Batch diversity (distinct-2) | 0.9367 |
 
 The teacher model string is the one the endpoint reported. It was reached
 through an OpenAI-compatible proxy, so it identifies the endpoint's advertised
@@ -88,59 +88,38 @@ model rather than an independently verified checkpoint.
 
 | `meaning` | Rows | Share |
 |---|---:|---:|
-| 0 | 143 | 9.1% |
-| 1 | 353 | 22.4% |
-| 2 | 114 | 7.2% |
-| 3 | 271 | 17.2% |
-| 4 | 697 | 44.2% |
+| 0 | 2017 | 16.7% |
+| 1 | 2611 | 21.6% |
+| 2 | 128 | 1.1% |
+| 3 | 2281 | 18.8% |
+| 4 | 5066 | 41.9% |
 
-Middle bands {1,2,3} hold **0.4394** of rows.
+Middle bands {1,2,3} hold **0.4269** of rows.
 
 ### Error tags
 
 | Tag | Count |
 |---|---:|
-| `art` | 617 |
-| `agr` | 428 |
-| `tense` | 402 |
-| `num` | 357 |
-| `prep` | 351 |
-| `form` | 274 |
-| `word` | 174 |
-| `pron` | 112 |
-| `other` | 96 |
-| `order` | 90 |
-| `punc` | 85 |
-| `coll` | 84 |
-| `poss` | 48 |
-| `unnat` | 47 |
-| `sp` | 39 |
-| `part` | 25 |
+| `art` | 4317 |
+| `prep` | 3018 |
+| `agr` | 2202 |
+| `tense` | 2110 |
+| `word` | 2055 |
+| `form` | 1905 |
+| `num` | 1800 |
+| `order` | 1355 |
+| `unnat` | 865 |
+| `coll` | 737 |
+| `pron` | 729 |
+| `poss` | 339 |
+| `sp` | 276 |
+| `punc` | 244 |
+| `part` | 162 |
+| `other` | 58 |
 
 ## Quality gates
 
-| Gate | Value | Threshold | Blocking | Result |
-|---|---:|---:|---|---|
-| `G1_self_consistency` | 0.9679 | 0.7 | yes | pass |
-| `G2_band_coverage` | 0.4677 | 0.4 | yes | pass |
-| `G3_format_validity` | 0.9645 | 0.9 | no | pass |
-| `G4_batch_diversity` | 0.9555 | 0.7 | no | pass |
-| `G5_other_tag_share` | 0.0297 | 0.05 | no | pass |
-| `G6_batch_single_parity` | 1.0 | 0.8 | no | pass |
-
-Overall: **pass**; blocking gates: **pass**.
-
-### Teacher self-consistency (the ceiling on any student)
-
-| Measure | Value |
-|---|---:|
-| `meaning` QWK | 0.967852 |
-| `correction` edit-F1 | 0.857129 |
-| Rows re-graded | 197 |
-
-The same sentences were graded twice, blind, with the cache disabled. A
-student cannot exceed the agreement its teacher has with itself, so report
-fidelity against these numbers rather than against 1.0.
+The pilot gate has not been run against this snapshot.
 
 ## Limitations
 
@@ -151,7 +130,6 @@ fidelity against these numbers rather than against 1.0.
 - **Bands are uncalibrated** until `lexi data calibrate` has run;
   `band_config.json` carries `"calibrated": false` when that is still true.
 - **`feedback` is unmeasured.** No metric in this snapshot evaluates it.
-- **The error mix is skewed relative to human-annotated learner text.** Against W&I+LOCNESS train split (18,224 rows, the same 16 tags), this data over-produces `agr` (13.2% vs 2.8%), `num` (11.1% vs 4.8%), `poss` (1.5% vs 0.7%) and under-produces `punc` (2.6% vs 15.4%), `sp` (1.2% vs 8.5%), `unnat` (1.5% vs 9.0%), `word` (5.4% vs 17.1%). The cause is the learner profiles: their `error_bias` fields drive what call 1 writes, and `punc`, `sp`, `pron`, `poss` and `other` appear in no profile at all. Punctuation and spelling are among the commonest real learner errors, so a model trained on this alone will be weakest there. Full comparison in `reports/tag-distribution-reference.json`.
 - **`grammar` and `naturalness` are uncalibrated.** `band_config.json` carries `"calibrated": false`, so those two bands come from the shipped design guesses rather than from this corpus's penalty distribution. `meaning` is unaffected — it is the teacher's own answer. Calibration on this corpus produced duplicate cut points, because 81% of rows carry no `usage` error at all and no threshold exists inside that mass of zeros; a five-band usage scale is not supported by data this clean.
 
 ## Not included