AvrovaDonz commited on
Commit
325ccfe
·
verified ·
1 Parent(s): a456e35

docs: align the README template with the published README wording

Browse files
components/google-math/tools/build_google_math.py CHANGED
@@ -930,10 +930,16 @@ school-level difficulty. Those pairs are **not research mathematics**, and the s
930
  projection of the paper's curriculum rather than the paper's dataset. Every record carries
931
  `content_origin` so consumers can separate verbatim source text from `generator_output`.
932
 
933
- **This corpus is mostly seeded generator output, not research mathematics.** `generated_qa_pair` records
934
- dominate the record count, the formal subset is a small minority, and **no Lean compilation and no proof
935
- verification was performed on any subset**, so no record here has a checked proof or a checked statement.
936
- `statistics.json` carries the measured counts, and the component README states them in full.
 
 
 
 
 
 
937
 
938
  ## Included sources
939
 
@@ -996,6 +1002,7 @@ The short version:
996
  measured yield.
997
  - `audit/excluded-duplicates.jsonl`, `audit/packing.json`, `audit/source-scan.json`.
998
  - `audit/validation-evidence.json`: the independent decode and round-trip evidence described below.
 
999
  - `statistics.json`: measured records, groups, splits, characters and packed counts, per source.
1000
  - `manifest.json`: SHA256 and byte size for every delivered file except itself and validation.json.
1001
  - `packed/`: mmla-type2-packed-v1 arrays, manifest, row attribution and per-record token counts.
@@ -1069,10 +1076,11 @@ boundary masks, and re-encodes every record's token stream against `data/*.jsonl
1069
  `all_packed_tokens_reencoded_source_shift_padding_masks_and_attribution`, `statistics_match_artifacts`).
1070
 
1071
  `tools/verify_google_math.py` is the independent check against the delivered files, and its output is
1072
- [`audit/validation-evidence.json`](audit/validation-evidence.json). It decoded 96 packed rows spread over
1073
- every shard of all three splits, and for each row (a) confirmed every row slice equals the re-encoding of
1074
- the corresponding span of `[BOS] + SentencePiece(text) + [EOS]` for its attributed source record and
1075
- (b) confirmed the decoded text equals the decoded source text modulo whitespace. It recorded 8 literal
 
1076
  unified diffs of a decoded row against its source record; every diff had 0 lines, whitespace identical.
1077
  It round-tripped 3 Parquet rows (indices 0, 239,658 and 479,315) back to their JSONL records with exact
1078
  match, and confirmed a `formal_lean_proof`, a `formal_lean_statement`, a `generated_qa_pair`, a
@@ -1080,31 +1088,34 @@ match, and confirmed a `formal_lean_proof`, a `formal_lean_statement`, a `genera
1080
  matching `text_sha256`. Split separation was recomputed from the data rather than trusted: 3,403 groups
1081
  in the audit and 3,403 in the data, no group spanning two splits, train/dev/test at 3,333/35/35 groups.
1082
 
1083
- [`audit/publish-check.json`](audit/publish-check.json) is a third check written by an independent publish
1084
- gate with its own code rather than this component's tools. It agrees with every total above (480,584
1085
- records, 3,403 groups, 78,364,888 characters), decodes 12 packed rows covering 248 whole-record segments
1086
- with 248/248 token-identical and whitespace-identical, matches 5/5 Parquet round-trips, and reports zero
1087
- groups spanning two splits.
 
 
 
 
 
1088
 
1089
  The record mix is deliberately stated because it is skewed: `generated_qa_pair` records are 99.31% of
1090
- records but 45.57% of characters, while formal Lean and prose records are 0.69% of records and 54.43% of
1091
- characters. A consumer wanting formal or prose mathematics should filter on `record_type` rather than
1092
- read the corpus whole.
1093
 
1094
  ## Publication
1095
 
1096
- The component is published as Hugging Face dataset pull request
1097
- [MMLA-ORG/Cayley-Math discussion #5](https://huggingface.co/datasets/MMLA-ORG/Cayley-Math/discussions/5),
1098
- on ref `refs/pr/5`, head commit `1894ffe2dc7f7f59abfb830f91e79738b1eb11d3`. It carries the whole
1099
- component plus the top-level `README.md` config and Collections update, and it has not been merged into
1100
- `main`.
1101
 
1102
  A token with `write` scope or org-admin rights is required to merge it. The token used for this build can
1103
  read the repository and open pull requests, but a direct commit to `main` is rejected with
1104
- `403 Forbidden: You have read access but not the required permissions for this operation`. Merging that
1105
  pull request, or re-running `hf upload` without `--create-pr` once such a token is available, completes
1106
- the submission. Earlier pull requests #3 and #4 are superseded by #5 and were left open rather than
1107
- closed, because closing them was outside this task's scope.
1108
 
1109
  ## Limitations
1110
 
 
930
  projection of the paper's curriculum rather than the paper's dataset. Every record carries
931
  `content_origin` so consumers can separate verbatim source text from `generator_output`.
932
 
933
+ **This corpus is mostly seeded generator output, not research mathematics.** Measured over the delivered
934
+ data, `generated_qa_pair` records are 477,276 of 480,584 records, which is 99.31% of records (0.993117)
935
+ and 35,710,731 of 78,364,888 text characters (45.57%). The formal subset is a small minority by record
936
+ count. Lean records total 3,027, namely 535 `formal_lean_proof` plus 2,492 `formal_lean_statement`, which
937
+ is 0.63% of records (0.006299). Geometry problems, prose proofs and notebooks add 281 more records, so
938
+ every non-generator record together is 3,308 records, 0.69% of the corpus (0.006883). Lean text does carry
939
+ 54.22% of the characters (0.542177) because Lean files are long, so a character-weighted view inverts the
940
+ balance while the record-weighted view does not. Read the composition from records, not from bytes.
941
+ **No Lean compilation and no proof verification was performed on any subset**, so no record in this corpus
942
+ has a checked proof or a checked statement.
943
 
944
  ## Included sources
945
 
 
1002
  measured yield.
1003
  - `audit/excluded-duplicates.jsonl`, `audit/packing.json`, `audit/source-scan.json`.
1004
  - `audit/validation-evidence.json`: the independent decode and round-trip evidence described below.
1005
+ - `audit/publish-check.json`: the second publish gate, measured from the delivered files after the build.
1006
  - `statistics.json`: measured records, groups, splits, characters and packed counts, per source.
1007
  - `manifest.json`: SHA256 and byte size for every delivered file except itself and validation.json.
1008
  - `packed/`: mmla-type2-packed-v1 arrays, manifest, row attribution and per-record token counts.
 
1076
  `all_packed_tokens_reencoded_source_shift_padding_masks_and_attribution`, `statistics_match_artifacts`).
1077
 
1078
  `tools/verify_google_math.py` is the independent check against the delivered files, and its output is
1079
+ [`audit/validation-evidence.json`](audit/validation-evidence.json). It decoded 96 packed rows, covering
1080
+ 68 of the 70 train shards plus the single dev and test shards, and for each row (a) confirmed every row
1081
+ slice equals the re-encoding of the corresponding span of `[BOS] + SentencePiece(text) + [EOS]` for its
1082
+ attributed source record and (b) confirmed the decoded text equals the decoded source text modulo
1083
+ whitespace. It recorded 8 literal
1084
  unified diffs of a decoded row against its source record; every diff had 0 lines, whitespace identical.
1085
  It round-tripped 3 Parquet rows (indices 0, 239,658 and 479,315) back to their JSONL records with exact
1086
  match, and confirmed a `formal_lean_proof`, a `formal_lean_statement`, a `generated_qa_pair`, a
 
1088
  matching `text_sha256`. Split separation was recomputed from the data rather than trusted: 3,403 groups
1089
  in the audit and 3,403 in the data, no group spanning two splits, train/dev/test at 3,333/35/35 groups.
1090
 
1091
+ [`audit/publish-check.json`](audit/publish-check.json) is a second publish gate with its own code rather
1092
+ than this component's tools. It agrees with every total above (480,584 records, 3,403 groups, 78,364,888
1093
+ characters, and the packed counts for all three splits), decodes 12 packed rows across all three splits
1094
+ and compares 263 attributed segments. All 263 row slices matched their canonical token slices by exact
1095
+ identity, and all 248 segments that cover a whole record decoded to text identical to the source record
1096
+ under whitespace normalisation and re-encoded to the identical token sequence. The other 15 segments are
1097
+ row-boundary fragments that begin or end inside a SentencePiece piece, so they cannot re-encode
1098
+ fragment-by-fragment; each one's decoded text was still located verbatim inside its attributed record.
1099
+ It matched 5 of 5 Parquet round-trips exactly and recomputed group separation from `data/*.jsonl.gz`
1100
+ directly, finding 0 groups spanning two splits.
1101
 
1102
  The record mix is deliberately stated because it is skewed: `generated_qa_pair` records are 99.31% of
1103
+ records but 45.57% of characters, while formal Lean records are 0.63% of records and 54.22% of characters,
1104
+ and all non-generator records together are 0.69% of records. A consumer wanting formal or prose
1105
+ mathematics should filter on `record_type` rather than read the corpus whole.
1106
 
1107
  ## Publication
1108
 
1109
+ The component is delivered as Hugging Face dataset pull request
1110
+ [MMLA-ORG/Cayley-Math discussion #3](https://huggingface.co/datasets/MMLA-ORG/Cayley-Math/discussions/3)
1111
+ on ref `refs/pr/3`. It carries the whole component plus the top-level `README.md` config and Collections
1112
+ update, and it has not been merged into `main`. Consult that discussion for the current merge state.
 
1113
 
1114
  A token with `write` scope or org-admin rights is required to merge it. The token used for this build can
1115
  read the repository and open pull requests, but a direct commit to `main` is rejected with
1116
+ `403 Forbidden: You have read access but not the required permissions for this operation`. Merging the
1117
  pull request, or re-running `hf upload` without `--create-pr` once such a token is available, completes
1118
+ the submission.
 
1119
 
1120
  ## Limitations
1121