Datasets:
id stringlengths 27 27 | text stringlengths 282 2.34k | task stringclasses 4
values | phase stringclasses 1
value | generator stringclasses 1
value | group_id stringlengths 17 21 | prompt_chars int64 170 1.62k | n_tokens int64 95 1.04k | text_sha256 stringlengths 64 64 | meta stringlengths 312 1.06k |
|---|---|---|---|---|---|---|---|---|---|
schema_map-90db7f708a2bc2c4 | Fill the target schema from this form. JSON object, schema keys in order, values copied verbatim, null when missing. Absent fields are null; values are copied as text exactly as written.
"""
Target schema (building permit):
permit_number, issue_date, site_address, applicant, contractor, work_type, valuation, fee_paid
... | form_map | stable | deterministic | schema|20261019|1280 | 537 | 298 | c2312e876cb419146106b81be5bc0e29d5d8b93bec2d21a19cb32d16f069ab89 | {"answer_form": "json", "frame_id": "form_map", "gold": "{\n \"permit_number\": \"BP744488\",\n \"issue_date\": null,\n \"site_address\": \"3534 Harbor Way, Stillwater Bend\",\n \"applicant\": \"Greta Northcott\",\n \"contractor\": null,\n \"work_type\": \"Roofing\",\n \"valuation\": \"$2,236,413.00\",\n \"fee_paid\": ... |
schema_map-a64374308f9816b7 | Map the form fields below to the schema and output JSON {field: value}; values unchanged, null for absent fields; ignore form fields outside the schema. Absent fields are null; values are copied as text exactly as written.
"""
Target schema (building permit):
permit_number, issue_date, site_address, applicant, contrac... | form_map | stable | deterministic | schema|20261019|8084 | 541 | 259 | 29757f131e33a5766ca5aaac825984e8bdbe8e41e7aaff5fedf676a8c8a9a489 | {"answer_form": "json", "frame_id": "form_map", "gold": "{\"permit_number\": \"BP710311\", \"issue_date\": null, \"site_address\": null, \"applicant\": \"Pembrook, Ines\", \"contractor\": \"Corvane Systems, Inc.\", \"work_type\": \"Roofing\", \"valuation\": \"$135,901.00\", \"fee_paid\": \"$1,589.00\"}", "marker": "wra... |
schema_map-e6407230252d90d9 | Spot the odd column out in the CSV below (name it, then one sentence of reasoning).
Excerpt:
Department,ISBN,Cost,Year,IN SERVICE,Unit #,Make
Public Works,978-310357674,"$51,425.00",2025,1,U1434,Chevrolet
Utilities,978-447567247,"$36,988.00",1994,1,U9914,Ford
Public Works,978-875050035,"$26,209.00",2000,1,U5048,Chevro... | odd_column | stable | deterministic | schema|20261019|9251 | 385 | 251 | 1d0e94b483d2aa436b87dd6ee0b5efc0f4216dd56cc70591a23000d64244a94f | {"answer_form": "prose", "frame_id": "odd_column", "gold": "\"ISBN\". Every other column is a fleet vehicle inventory attribute; \"ISBN\" comes from a library catalog record.", "marker": "wrap:5", "n_cols": 7, "n_extra": 1, "n_missing": 1, "presentation": "plain", "source": "synthetic-typed-record", "stratum": "schema:... |
schema_map-d7ebd8b719663612 | Task: Produce the target-schema version of this CSV: rename and reorder the columns to match the schema, keep every value as written, add empty columns for missing fields, and omit columns the schema does not define. Leave missing fields empty and drop columns the schema does not define.
Passage:
Target schema (buildi... | remap_csv | stable | deterministic | schema|20261019|14097 | 1,232 | 656 | 92068dfba1e6b2ca2e0e9dfc1cfda81a28595d92ca036df018918d421ddf03df | {"answer_form": "csv", "frame_id": "remap_csv", "gold": "permit_number,issue_date,site_address,applicant,contractor,work_type,valuation,fee_paid\nBP331517,,,Verna Quarles,Driftwood Capital LP,Demolition,1618078.00,\nBP534721,,,Dorian Ivers,Fenwick Grain Co.,Addition,1460817.00,\nBP813075,,,Edwina Eastwick,Bluefen Logis... |
schema_map-1f2055d40d043767 | Extract the filled-in values of this form into JSON keyed by the target schema fields. Use null for schema fields the form does not include. Copy values exactly as written. Absent fields are null; values are copied as text exactly as written.
Excerpt:
Target schema (court case record):
case_number, court, filing_date,... | form_map | stable | deterministic | schema|20261019|21374 | 509 | 221 | fbb7e2f7a2cfc86cab3de8d54af1fb5d0ba6cd3e507c43467155eb249bff6577 | {"answer_form": "json", "frame_id": "form_map", "gold": "{\n \"case_number\": \"CV-902331\",\n \"court\": \"Superior Court\",\n \"filing_date\": null,\n \"judge\": \"Lockhart, Edwina\",\n \"plaintiff\": \"Bluefen Logistics Corp.\",\n \"defendant\": null,\n \"case_type\": null,\n \"status\": \"Dismissed\"\n}", "marker":... |
schema_map-0f72d1b30a5326d2 | <|start|><|user|>
Spot the odd column out in the CSV below (name it, then one sentence of reasoning).
---
agcy,FR Doc No,Type,Comment Period Ends,Pub Date,Batting Average
Environmental Protection Agency,2019-68730,Proposed Rule,05/12/2021,02/04/2021,46.6%
Coast Guard,2019-72278,Notice,11/28/2013,09/05/2013,66.7%
Food ... | odd_column | stable | deterministic | schema|20261019|15217 | 422 | 211 | b8dee4ef84d8a2adc6fc99d5a27a6c9e50caeb03d0bb53ce1c1fcfc7e527e4ae | {"answer_form": "prose", "frame_id": "odd_column", "gold": "\"Batting Average\" does not belong: it is a sports statistics field, while the other columns describe a Federal Register document index.", "marker": "wrap:1", "n_cols": 6, "n_extra": 1, "n_missing": 1, "presentation": "chat", "source": "synthetic-typed-record... |
schema_map-97f88515a03d398d | Request: Map the columns of this CSV extract to the target schema below. Return JSON with "mapping" (each source column -> target field, or null if it has no target) and "missing" (target fields with no source column). Columns that match no target field map to null.
---
Target schema (grant award):
{
"grant_number": ... | map_json | stable | deterministic | schema|20261019|6964 | 1,017 | 415 | 82a42fb24687fa4d0bae94abb1841a349ae2a1c19d024f7dc2aa2ef79f068d12 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\"mapping\": {\"City\": \"city\", \"grant_num\": \"grant_number\", \"Rcpt Name\": \"recipient\", \"Award Amount\": \"amount_awarded\", \"principal_investigator\": \"project_director\"}, \"missing\": [\"program\", \"fiscal_year\"]}", "marker": "wrap:1|Request:", ... |
schema_map-8f0305c55ada7c77 | <|start|><|user|>
Convert these labeled form entries into a JSON record that follows the schema below (null for fields the form lacks; extra form fields are dropped). Absent fields are null; values are copied as text exactly as written.
Passage:
Target schema (food-safety inspection):
inspection_id, establishment, ins... | form_map | stable | deterministic | schema|20261019|1708 | 610 | 258 | 2e6a7d16da51792220409b92abf37ed97f0fc03ee5e617ab823fcdaa3e005029 | {"answer_form": "json", "frame_id": "form_map", "gold": "{\"inspection_id\": \"INS1837301\", \"establishment\": \"Elmstead Holdings\", \"inspection_date\": \"09/03/2016\", \"inspector\": \"Northcott, Nolan\", \"score\": \"58\", \"result\": \"Pass\", \"critical_violations\": \"1\"}", "marker": "wrap:3", "n_cols": 9, "n_... |
schema_map-986826a96af7934b | Spot the odd column out in the CSV below (name it, then one sentence of reasoning).
Passage:
YR MODEL,Make,unit_num,purchase_price,Blood Type,Assigned Dept
2010,Ram,U2031,73538.00,B+,Police
1997,Chevrolet,U2618,87191.00,O-,Parks
2012,Ram,U9382,65382.00,A+,Parks
"Blood Type". Every other column is a fleet vehicle inve... | odd_column | stable | deterministic | schema|20261019|19339 | 264 | 163 | a2258dcaf183cb9fe487943c5f1cdad358a5a7bf5359e3a7718a655c389ab445 | {"answer_form": "prose", "frame_id": "odd_column", "gold": "\"Blood Type\". Every other column is a fleet vehicle inventory attribute; \"Blood Type\" comes from a medical record.", "marker": "wrap:3", "n_cols": 6, "n_extra": 1, "n_missing": 2, "presentation": "plain", "source": "synthetic-typed-record", "stratum": "sch... |
schema_map-21ca85e204205f55 | <|start|><|user|>
Convert the extract below into a CSV that follows the Federal Register document index schema exactly (same field names and order). Do not reformat any value; leave missing fields empty and drop extra columns. Fields the source lacks stay empty, and source columns outside the schema are dropped.
---
T... | remap_csv | stable | deterministic | schema|20261019|4182 | 1,008 | 445 | 5f1bed3330c58136d32a7365d0ccf0dbd5702a35413f8e1886eb2d9c6b7010b6 | {"answer_form": "csv", "frame_id": "remap_csv", "gold": "document_number,agency,document_type,publication_date,comment_deadline,page_count\n2019-11349,Environmental Protection Agency,Proposed Rule,23-Jun-2006,15-Dec-2006,135\n2019-66579,Forest Service,Notice,18-Jan-2020,28-Jan-2021,16\n2019-81489,Food and Drug Administ... |
schema_map-853b0fbc5af2bbf4 | Align these source columns with the food-safety inspection schema. Output a JSON object whose "mapping" keys are the source headers exactly as written, plus a "missing" list. Columns that match no target field map to null.
---
Target schema (food-safety inspection):
inspection_id, establishment, inspection_date, inspe... | map_json | stable | deterministic | schema|20261019|14320 | 600 | 319 | bb848be3da449326546ffcfb84a75f69a0e3cc02372874d1930df551545ea3f8 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\n \"mapping\": {\n \"Insepcted By\": \"inspector\",\n \"INSPECTION SCORE\": \"score\",\n \"Insp #\": \"inspection_id\",\n \"Insp Dt\": \"inspection_date\"\n },\n \"missing\": [\n \"establishment\",\n \"result\",\n \"critical_violations\"\n ]\n}", "marker... |
schema_map-416d56614791556d | <|start|><|user|>
Convert the extract below into a CSV that follows the property tax assessment schema exactly (same field names and order). Do not reformat any value; leave missing fields empty and drop extra columns. Fields the source lacks stay empty, and source columns outside the schema are dropped.
<text>
Target... | remap_csv | stable | deterministic | schema|20261019|9707 | 1,139 | 715 | da4b8ba0d9cab707b9f3bb39daa99f948b15b91b7b479794c21800b895e7689c | {"answer_form": "csv", "frame_id": "remap_csv", "gold": "parcel_id,owner,situs_address,tax_year,land_value,improvement_value,exempt\nP-62414626,Bertram Abernathy,\"788 Orchard Dr, Marlow Heights\",2020,\"$238,514.00\",,No\nP-53879223,Quinn Lockhart,\"6620 Mill Ave, Juniper Flats\",1990,\"$471,991.00\",,No\nP-56752828,O... |
schema_map-d198594b06dac600 | Q: Schema mapping task: for every column in the data, give the target field it corresponds to (null for columns the schema does not have), then list the schema fields the data lacks. JSON only. Columns that match no target field map to null.
```
Target schema (food-safety inspection):
{
"inspection_id": "string (iden... | map_json | stable | deterministic | schema|20261019|9019 | 852 | 357 | 975b0599130b3994c57344590bf985c4cd8f63a3cc59c732dd19a9a6a78c4ad7 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\"mapping\": {\"INSP RESULT\": \"result\", \"Points\": \"score\", \"Insp #\": \"inspection_id\", \"Notes\": null, \"Date Inspected\": \"inspection_date\", \"Facility Name\": \"establishment\", \"Sanitarian\": \"inspector\", \"Critical Violations\": \"critical_vi... |
schema_map-5169abd64b05b969 | Instruction: This grant award form uses its own field labels. Return a JSON object with the target schema's keys and the form's values (exact text), null where the form has no such field. Absent fields are null; values are copied as text exactly as written.
Text:
Target schema (grant award):
grant_number, recipient, p... | form_map | stable | deterministic | schema|20261019|4117 | 559 | 248 | 36d0da0e5472d66da882ab71b434d814b0ce160ac69cde29970356b79bddf75a | {"answer_form": "json", "frame_id": "form_map", "gold": "{\"grant_number\": \"G-751721\", \"recipient\": \"Juniper Ridge Energy\", \"program\": \"Workforce Training\", \"fiscal_year\": \"1994\", \"amount_awarded\": \"1362980.00\", \"project_director\": \"Wendell Northcott\", \"city\": \"Ashby Crossing\"}", "marker": "w... |
schema_map-dbac87ca064dd434 | Request: The header row below is messy (abbreviations, typos, different order). Match each source column to a field of the target schema and report which target fields are not present. Answer as JSON: {"mapping": {...}, "missing": [...]}. Columns that match no target field map to null.
Text:
Target schema (food-safety... | map_json | stable | deterministic | schema|20261019|18487 | 693 | 338 | 362af8f1c087f53c29e195b277df8fc298c5ff81f9267780d21c062aa42e78ef | {"answer_form": "json", "frame_id": "map_json", "gold": "{\n \"mapping\": {\n \"INSPECTION DATE\": \"inspection_date\",\n \"INSPECTION ID\": \"inspection_id\",\n \"Isnp Name\": \"inspector\",\n \"Priority Violations\": \"critical_violations\",\n \"Outcome\": \"result\",\n \"Inspection Scoer\": \"score\"\n },\n \"... |
schema_map-80996a7576bc3045 | Which header doesn't fit this data set? Name it and give the reason in one sentence.
Text:
Grantee,Recipient City,Shoe Size,Grant No,FundingProgram,PrincipalInvestigator
Hollowell Medical Group,Juniper Flats,1839,G-158774,Rural Water,"Pembrook, Verna"
Driftwood Capital LP,Ashby Crossing,2743,G-784441,Historic Preserva... | odd_column | stable | deterministic | schema|20261019|1863 | 610 | 264 | 48c61b337d49a123252a8871d18da5d159be84402af677754a53dfec5dae3416 | {"answer_form": "prose", "frame_id": "odd_column", "gold": "\"Shoe Size\" does not belong: it is a retail inventory field, while the other columns describe a grant award.", "marker": "wrap:2", "n_cols": 6, "n_extra": 1, "n_missing": 2, "presentation": "plain", "source": "synthetic-typed-record", "stratum": "schema:gran... |
schema_map-2ae45b51358ed28e | Normalize the headers of this Federal Register document index file to the schema below and say which fields are absent. Return {"mapping": {source_header: target_field_or_null}, "missing": [target_field, ...]}. Columns that match no target field map to null.
"""
Target schema (Federal Register document index):
documen... | map_json | stable | deterministic | schema|20261019|17596 | 561 | 262 | 31893d1d85adc22ccc2b157a96e34cd9a197d604dc709f4f4cd1664ef10bec71 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\n \"mapping\": {\n \"PUBLISHED\": \"publication_date\",\n \"Action Type\": \"document_type\",\n \"Comments Due\": \"comment_deadline\",\n \"PG CT\": \"page_count\",\n \"Agcy\": \"agency\"\n },\n \"missing\": [\n \"document_number\"\n ]\n}", "marker": "wra... |
schema_map-6b74356e40650c2f | ### Task
The header row below is messy (abbreviations, typos, different order). Match each source column to a field of the target schema and report which target fields are not present. Answer as JSON: {"mapping": {...}, "missing": [...]}. Columns that match no target field map to null.
"""
Target schema (Federal Regis... | map_json | stable | deterministic | schema|20261019|14124 | 667 | 310 | 276139f7919a7d744143aac8c65c710c2887c04baffe77fb03ee9976c25b9fac | {"answer_form": "json", "frame_id": "map_json", "gold": "{\n \"mapping\": {\n \"Issuing Aegncy\": \"agency\",\n \"Pages\": \"page_count\",\n \"CATEGORY\": \"document_type\",\n \"comments_due\": \"comment_deadline\",\n \"PubDt\": \"publication_date\"\n },\n \"missing\": [\n \"document_number\"\n ]\n}", "marker": "... |
schema_map-27ab65626604276c | <|start|><|user|>
Which header doesn't fit this data set? Name it and give the reason in one sentence.
Passage:
Veh Unit,year,Make,Tide Height (ft),SERVICE STATUS FLAG,Cost
U6828,2002,Ram,690,N,"$83,956.00"
U1429,2019,Ford,4380,N,"$86,752.00"
U9450,2018,Toyota,2566,N,"$47,894.00"<|eot|>
<|assistant|>
The odd column is... | odd_column | stable | deterministic | schema|20261019|10924 | 303 | 181 | bcc45881a365785fad1cf4d635adeeb9d7cc28dfc904f74aff68c8c0b2ea8917 | {"answer_form": "prose", "frame_id": "odd_column", "gold": "The odd column is \"Tide Height (ft)\" — tide table data has nothing to do with a fleet vehicle inventory.", "marker": "wrap:3", "n_cols": 6, "n_extra": 1, "n_missing": 2, "presentation": "chat", "source": "synthetic-typed-record", "stratum": "schema:vehicle",... |
schema_map-7c8d411f6ee4acef | <|start|><|user|>
Align these source columns with the grant award schema. Output a JSON object whose "mapping" keys are the source headers exactly as written, plus a "missing" list. Columns that match no target field map to null.
Text:
Target schema (grant award):
- grant_number (string (identifier))
- recipient (stri... | map_json | stable | deterministic | schema|20261019|18942 | 844 | 342 | 1678d15edf721c25f9a235cfe28a5b44de1aa4006c9931cfa6b909cd9e312b3d | {"answer_form": "json", "frame_id": "map_json", "gold": "{\"mapping\": {\"FUNDING PROGRAM\": \"program\", \"AMT AWARDED\": \"amount_awarded\", \"grant_no\": \"grant_number\", \"Recipient City\": \"city\", \"Ficsal Year\": \"fiscal_year\", \"Recipient Organization\": \"recipient\"}, \"missing\": [\"project_director\"]}"... |
schema_map-b423aef3bf36c6a5 | Find the column that is out of place in this table and explain briefly.
---
Dkt oN,Status,Plaintiff,Jdg,date_filed,Defendatn,TYPE OF CASE,Court,Seating Capacity
CV-760111,Closed,Driftwood Capital LP,Fletcher Quarles,7/25/21,Lionel Calloway,Property,Municipal Court,3567
CV-164730,Open,Elmstead Holdings,Wendell Calloway... | odd_column | stable | deterministic | schema|20261019|23611 | 591 | 285 | a169a4943ae1618b9c79280fcf4e9b4b32ea944e34012be80f03e51a2ff81cfb | {"answer_form": "prose", "frame_id": "odd_column", "gold": "\"Seating Capacity\". Every other column is a court case record attribute; \"Seating Capacity\" comes from a venue listing record.", "marker": "wrap:1", "n_cols": 9, "n_extra": 1, "n_missing": 0, "presentation": "plain", "source": "synthetic-typed-record", "st... |
schema_map-5bd88fb9525fa948 | <|start|><|user|>
Re-map the columns of this file onto the target schema and output the resulting CSV (header = schema fields in order). Values stay as they are. Leave missing fields empty and drop columns the schema does not define.
```
Target schema (building permit):
permit_number, issue_date, site_address, applica... | remap_csv | stable | deterministic | schema|20261019|21373 | 919 | 629 | e216cc9f049a82b0c85dbcfa3889093fb340e0e1bdebd3f27da792b388c1699d | {"answer_form": "csv", "frame_id": "remap_csv", "gold": "permit_number,issue_date,site_address,applicant,contractor,work_type,valuation,fee_paid\n,8/1/11,\"4081 Commerce Blvd, Stillwater Bend\",,Elmstead Holdings,Alteration,313651.00,92.00\n,4/9/19,\"6706 Ridge Ln, Ashby Crossing\",,Juniper Ridge Energy,Roofing,57234.0... |
schema_map-0c869c80d8aad18b | Request: Map the form fields below to the schema and output JSON {field: value}; values unchanged, null for absent fields; ignore form fields outside the schema. Absent fields are null; values are copied as text exactly as written.
Text:
Target schema (court case record):
case_number, court, filing_date, judge, plaint... | form_map | stable | deterministic | schema|20261019|11638 | 514 | 218 | 816d1fb747316f4afbbd8b619f5b26db0b7baac872d06e2c3ea74cbadda0a8db | {"answer_form": "json", "frame_id": "form_map", "gold": "{\"case_number\": \"CV-647835\", \"court\": \"Superior Court\", \"filing_date\": null, \"judge\": null, \"plaintiff\": \"Halvorsen Industrial LLC\", \"defendant\": \"Nolan Dunmore\", \"case_type\": \"Tort\", \"status\": \"Dismissed\"}", "marker": "wrap:2|Request:... |
schema_map-00007a67caff7071 | Q: Map and reorder: give the CSV rewritten to the government contract award schema (schema field order), copying each value exactly; fields with no source column are empty; unmatched source columns are left out. Fields the source lacks stay empty, and source columns outside the schema are dropped.
```
Target schema (g... | remap_csv | stable | deterministic | schema|20261019|18176 | 793 | 500 | 5f9bc992f73cc0440ac924f316d46d1b6c32d04afa38b59728eaebc2d91abba6 | {"answer_form": "csv", "frame_id": "remap_csv", "gold": "contract_id,vendor,award_date,start_date,end_date,amount_usd,agency,set_aside\nGS-8557056,,1998-07-02,,,\"$926,984.00\",Dept. of Transportation,\nGS-7380829,,2017-12-17,,,\"$1,295,998.00\",Dept. of Transportation,\nGS-2505231,,2011-04-16,,,\"$1,758,858.00\",Dept.... |
schema_map-3f2824e084096359 | <|start|><|user|>
Map the columns of this CSV extract to the target schema below. Return JSON with "mapping" (each source column -> target field, or null if it has no target) and "missing" (target fields with no source column). Columns that match no target field map to null.
Excerpt:
Target schema (fleet vehicle inven... | map_json | stable | deterministic | schema|20261019|16636 | 918 | 447 | 06bb26eb0b5d4d68055ed52c9bdb8ad69cbb411b5fa8106b200ebabce3a40e62 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\n \"mapping\": {\n \"in_service\": \"in_service\",\n \"Vehicle Make\": \"make\",\n \"AssignedDept\": \"department\",\n \"Batch\": null,\n \"Model Year\": \"model_year\",\n \"unit_no\": \"unit_number\",\n \"Miles\": \"odometer\",\n \"Notes\": null\n },\n... |
schema_map-36bcea458fcfea38 | Query: Map the columns of this CSV extract to the target schema below. Return JSON with "mapping" (each source column -> target field, or null if it has no target) and "missing" (target fields with no source column). Columns that match no target field map to null.
---
Target schema (fleet vehicle inventory):
{
"unit_... | map_json | stable | deterministic | schema|20261019|22059 | 787 | 356 | d128c54192202b2613de25d4408a5e7bde05f9f0a7bd00a270731deac58e06b3 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\"mapping\": {\"MY\": \"model_year\", \"In Svc\": \"in_service\", \"Unit No\": \"unit_number\", \"Odometer\": \"odometer\", \"using_department\": \"department\"}, \"missing\": [\"make\", \"purchase_price\"]}", "marker": "wrap:1|Query:", "n_cols": 5, "n_extra": 0... |
schema_map-aaf3b3816b6b7441 | <|start|><|user|>
Rewrite this CSV to the target schema: columns in schema order with the schema's field names, values copied unchanged, empty cells for fields the source lacks, and source columns outside the schema dropped. Leave missing fields empty and drop columns the schema does not define.
<text>
Target schema (... | remap_csv | stable | deterministic | schema|20261019|13967 | 936 | 550 | 1cdd07b252c7d1400623f3e9f4386749dfa4cbb3db4a2e2c4e5287bd6d97e899 | {"answer_form": "csv", "frame_id": "remap_csv", "gold": "contract_id,vendor,award_date,start_date,end_date,amount_usd,agency,set_aside\n,Fenwick Grain Co.,05/11/2017,10/31/2017,06/27/2018,\"$8,267,905.00\",General Services Administration,Yes\n,Granite Arch Partners,02/01/2022,06/28/2022,04/25/2023,\"$1,025,168.00\",Dep... |
schema_map-518adb9e8eb2de2b | One column in this fleet vehicle inventory extract does not belong with the others. Which one is it, and why? Answer in one sentence that names the column.
---
Blood Type,Cost,make,assigned_dept,Year,fleet_number,Active?
AB+,"$39,833.00",Chevrolet,Parks,2016,U8192,0
A+,"$59,728.00",Toyota,Utilities,2004,U2219,1
B+,"$2... | odd_column | stable | deterministic | schema|20261019|5785 | 362 | 182 | 6648e4c731941db9d055b55e07e44477b1da9ce7614c58d042225365957a04cc | {"answer_form": "prose", "frame_id": "odd_column", "gold": "The odd column is \"Blood Type\" — medical data has nothing to do with a fleet vehicle inventory.", "marker": "wrap:1", "n_cols": 7, "n_extra": 1, "n_missing": 1, "presentation": "plain", "source": "synthetic-typed-record", "stratum": "schema:vehicle", "typo":... |
schema_map-cc6954921db21f75 | <|start|><|user|>
Spot the odd column out in the CSV below (name it, then one sentence of reasoning).
<text>
Property Location,Batting Average,Yr,OWNR,Ldn Val,Bldg Vlaue
"6578 Elm St, Wexmoor",42.1%,2016,"Brightwater, Jasper","$539,672.00","$1,858,635.00"
"1674 Mill Ave, Copper Mesa",12.8%,2004,"Merriweather, Kendra",... | odd_column | stable | deterministic | schema|20261019|3792 | 551 | 318 | 0052a9d1a7aed084a3fd93237955315d99c8a5485a773311925a93fac68eb215 | {"answer_form": "prose", "frame_id": "odd_column", "gold": "\"Batting Average\". Every other column is a property tax assessment attribute; \"Batting Average\" comes from a sports statistics record.", "marker": "wrap:4", "n_cols": 6, "n_extra": 1, "n_missing": 2, "presentation": "chat", "source": "synthetic-typed-recor... |
schema_map-16fd16fe9c3a368c | <|start|><|user|>
Re-map the columns of this file onto the target schema and output the resulting CSV (header = schema fields in order). Values stay as they are. Fields the source lacks stay empty, and source columns outside the schema are dropped.
---
Target schema (government contract award):
- contract_id (string (... | remap_csv | stable | deterministic | schema|20261019|16664 | 1,083 | 647 | 21259817eeee8ddf51cd4d2e4f0d8ff684374ed050f183d3f027ae1a2443708a | {"answer_form": "csv", "frame_id": "remap_csv", "gold": "contract_id,vendor,award_date,start_date,end_date,amount_usd,agency,set_aside\nGS-9225120,\"Corvane Systems, Inc.\",02/17/2020,12/08/2020,10/18/2021,2148814.00,Dept. of Transportation,1\nGS-7424608,Elmstead Holdings,03/23/2016,04/21/2016,12/04/2016,5217258.00,Dep... |
schema_map-0995ee2aeb1e3b4a | <|start|><|user|>
Extract the filled-in values of this form into JSON keyed by the target schema fields. Use null for schema fields the form does not include. Copy values exactly as written. Absent fields are null; values are copied as text exactly as written.
---
Target schema (property tax assessment):
parcel_id, ow... | form_map | stable | deterministic | schema|20261019|17838 | 575 | 297 | daa2ee41819313d9405dc6ec81d5426e5f8ed577a35467dfddab7c551440e9d4 | {"answer_form": "json", "frame_id": "form_map", "gold": "{\"parcel_id\": \"P-97692262\", \"owner\": \"Hargreave, Celia\", \"situs_address\": \"5152 Ridge Ln, Wexmoor\", \"tax_year\": \"2002\", \"land_value\": \"709636.00\", \"improvement_value\": \"169101.00\", \"exempt\": null}", "marker": "wrap:1", "n_cols": 7, "n_ex... |
schema_map-fd0fd19e79eab0a6 | Convert these labeled form entries into a JSON record that follows the schema below (null for fields the form lacks; extra form fields are dropped). Absent fields are null; values are copied as text exactly as written.
```
Target schema (Federal Register document index):
{
"document_number": "string (identifier)",
"... | form_map | stable | deterministic | schema|20261019|14850 | 767 | 280 | 5d155296051c0bc49dd3593949631b807894d494c59175f2ea14865a2206e9b6 | {"answer_form": "json", "frame_id": "form_map", "gold": "{\"document_number\": \"2019-44971\", \"agency\": \"Food and Drug Administration\", \"document_type\": \"Notice\", \"publication_date\": \"August 14, 2012\", \"comment_deadline\": \"November 22, 2012\", \"page_count\": null}", "marker": "wrap:6", "n_cols": 6, "n_... |
schema_map-a038721c388573fb | Request: Map the columns of this CSV extract to the target schema below. Return JSON with "mapping" (each source column -> target field, or null if it has no target) and "missing" (target fields with no source column). Columns that match no target field map to null.
Passage:
Target schema (Federal Register document in... | map_json | stable | deterministic | schema|20261019|7320 | 844 | 421 | a7f880e54bf0c27c51016558d2feafcdb179a3b7de226dc06732f93dbbf3e669 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\n \"mapping\": {\n \"PAGE COUNT\": \"page_count\",\n \"AGENCY NAME\": \"agency\",\n \"FRDOC\": \"document_number\",\n \"Batch\": null,\n \"Published\": \"publication_date\",\n \"Action Type\": \"document_type\",\n \"Ctm Deadline\": \"comment_deadline\"\n... |
schema_map-b827ba5dec30a007 | Question: Schema mapping task: for every column in the data, give the target field it corresponds to (null for columns the schema does not have), then list the schema fields the data lacks. JSON only. Columns that match no target field map to null.
```
Target schema (court case record):
- case_number (string (identifi... | map_json | stable | deterministic | schema|20261019|11259 | 1,176 | 488 | 97ffc20b7326c4ce48b2a6ad3fef05bfe767e8127a87c1d7a5753afb4b532498 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\n \"mapping\": {\n \"Case Num\": \"case_number\",\n \"COURT\": \"court\",\n \"FILED ON\": \"filing_date\",\n \"petitioner_plaintiff\": \"plaintiff\",\n \"nature_of_suit\": \"case_type\",\n \"Assigned Judge\": \"judge\",\n \"Dft Name\": \"defendant\"\n },... |
schema_map-fd538319362dc30c | [Task]
Re-map the columns of this file onto the target schema and output the resulting CSV (header = schema fields in order). Values stay as they are. Fields the source lacks stay empty, and source columns outside the schema are dropped.
Excerpt:
Target schema (fleet vehicle inventory):
{
"unit_number": "string (iden... | remap_csv | stable | deterministic | schema|20261019|6660 | 729 | 324 | 426804d6e1eced3a1c343f2aec5d051edf6d919ff0a9c6665324c208ef451a42 | {"answer_form": "csv", "frame_id": "remap_csv", "gold": "unit_number,model_year,make,odometer,department,purchase_price,in_service\n,2006,,94142,Public Works,\"$43,078.00\",Y\n,2017,,72127,Public Works,\"$26,605.00\",N", "marker": "wrap:5|[Task]", "n_cols": 5, "n_extra": 0, "n_missing": 2, "presentation": "qa", "source... |
schema_map-89bfd30eabb0064e | <|start|><|user|>
Fill the target schema from this form. JSON object, schema keys in order, values copied verbatim, null when missing. Absent fields are null; values are copied as text exactly as written.
Excerpt:
Target schema (fleet vehicle inventory):
unit_number, model_year, make, odometer, department, purchase_pr... | form_map | stable | deterministic | schema|20261019|6108 | 487 | 208 | 143cc005c14e29a3d2ac95dafda9326617686e5e7f9bf0df0ca1a223215a5ced | {"answer_form": "json", "frame_id": "form_map", "gold": "{\"unit_number\": null, \"model_year\": null, \"make\": null, \"odometer\": \"165317\", \"department\": \"Utilities\", \"purchase_price\": \"80415.00\", \"in_service\": \"FALSE\"}", "marker": "wrap:5", "n_cols": 6, "n_extra": 2, "n_missing": 3, "presentation": "c... |
schema_map-0ed3adb725d41935 | Align these source columns with the building permit schema. Output a JSON object whose "mapping" keys are the source headers exactly as written, plus a "missing" list. Columns that match no target field map to null.
---
Target schema (building permit):
- permit_number (string (identifier))
- issue_date (date)
- site_a... | map_json | stable | deterministic | schema|20261019|7631 | 963 | 426 | e5bb3c758f68e90720b635efaebb06cdaff9da26fbf11844a1478c0ecdac3eb6 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\n \"mapping\": {\n \"Requested By\": \"applicant\",\n \"EST. VALUE\": \"valuation\",\n \"PROPERTY ADDRESS\": \"site_address\",\n \"Fee Amt\": \"fee_paid\",\n \"scope\": \"work_type\",\n \"LicensedContractor\": \"contractor\"\n },\n \"missing\": [\n \"per... |
schema_map-0c2f104f5592154d | <|start|><|user|>
Reshape this data to the schema below. Output CSV only, header first, with values unchanged. Fields the source lacks stay empty, and source columns outside the schema are dropped.
Passage:
Target schema (government contract award):
- contract_id (string (identifier))
- vendor (string (organization))
... | remap_csv | stable | deterministic | schema|20261019|1407 | 802 | 371 | 4d8169d4cb38a27f354424aa4a7f1c50d6ee46065ddfb22b9181f766ffd53cc4 | {"answer_form": "csv", "frame_id": "remap_csv", "gold": "contract_id,vendor,award_date,start_date,end_date,amount_usd,agency,set_aside\nGS-9882975,Granite Arch Partners,,,11/11/99,5445361.00,General Services Administration,1\nGS-9389415,Hollowell Medical Group,,,12/3/23,8242890.00,Dept. of Transportation,0", "marker": ... |
schema_map-353287dff0f9331a | Produce the target-schema version of this CSV: rename and reorder the columns to match the schema, keep every value as written, add empty columns for missing fields, and omit columns the schema does not define. Leave missing fields empty and drop columns the schema does not define.
```
Target schema (grant award):
gra... | remap_csv | stable | deterministic | schema|20261019|14308 | 1,021 | 696 | ca3db575062e916159d4056b9380cf8355b675de599a2288e6a935a5e5c0a914 | {"answer_form": "csv", "frame_id": "remap_csv", "gold": "grant_number,recipient,program,fiscal_year,amount_awarded,project_director,city\nG-338811,Juniper Ridge Energy,Broadband Access,2013,\"$1,233,107.00\",Quinn Calloway,Brindle Point\nG-919772,Halvorsen Industrial LLC,Workforce Training,2003,\"$1,567,760.00\",Jasper... |
schema_map-4c431be6f0d2456f | <|start|><|user|>
Map the columns of this CSV extract to the target schema below. Return JSON with "mapping" (each source column -> target field, or null if it has no target) and "missing" (target fields with no source column). Columns that match no target field map to null.
Excerpt:
Target schema (fleet vehicle inven... | map_json | stable | deterministic | schema|20261019|20595 | 550 | 247 | dc5be832520a1bc2b600ac340357165c87120615956099717f2dd71fad41f7ee | {"answer_form": "json", "frame_id": "map_json", "gold": "{\"mapping\": {\"acq_cost\": \"purchase_price\", \"MANUFACTURER\": \"make\", \"MY\": \"model_year\", \"DEPT\": \"department\", \"MILEAGE\": \"odometer\"}, \"missing\": [\"unit_number\", \"in_service\"]}", "marker": "wrap:5", "n_cols": 5, "n_extra": 0, "n_missing"... |
schema_map-b22e6a51274a0026 | <|start|><|system|>
You are a precise assistant for text and documents.<|eot|>
<|user|>
Read the form and populate the government contract award schema as JSON. Do not reformat dates or amounts. Absent fields are null; values are copied as text exactly as written.
<text>
Target schema (government contract award):
cont... | form_map | stable | deterministic | schema|20261019|1679 | 534 | 233 | 7014a0ad372b67913a4f2684263b9f100e729eb7aaae36e02e6b472d19398a5b | {"answer_form": "json", "frame_id": "form_map", "gold": "{\"contract_id\": \"GS-4434922\", \"vendor\": null, \"award_date\": \"03/29/2000\", \"start_date\": null, \"end_date\": null, \"amount_usd\": null, \"agency\": null, \"set_aside\": \"FALSE\"}", "marker": "wrap:4", "n_cols": 4, "n_extra": 1, "n_missing": 5, "prese... |
schema_map-3e63300e2773cf0e | Extract the filled-in values of this form into JSON keyed by the target schema fields. Use null for schema fields the form does not include. Copy values exactly as written. Absent fields are null; values are copied as text exactly as written.
Passage:
Target schema (property tax assessment):
parcel_id, owner, situs_ad... | form_map | stable | deterministic | schema|20261019|19046 | 564 | 305 | d4aaebabf477dfab477907846f9cf8dc4bb6c3f9216c6148ce7ea8afd723e18b | {"answer_form": "json", "frame_id": "form_map", "gold": "{\n \"parcel_id\": \"P-31245752\",\n \"owner\": \"Redfern, Jasper\",\n \"situs_address\": \"7727 Commerce Blvd, Wexmoor\",\n \"tax_year\": null,\n \"land_value\": \"$535,737.00\",\n \"improvement_value\": \"$101,726.00\",\n \"exempt\": \"1\"\n}", "marker": "wrap:... |
schema_map-232d9dd1b9fcc1cf | <|start|><|user|>
Normalize the headers of this food-safety inspection file to the schema below and say which fields are absent. Return {"mapping": {source_header: target_field_or_null}, "missing": [target_field, ...]}. Columns that match no target field map to null.
"""
Target schema (food-safety inspection):
{
"ins... | map_json | stable | deterministic | schema|20261019|11207 | 1,035 | 439 | a924faaf28135f403b515389dcf341af9271671ea73311016d4d85ddcf60ce81 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\n \"mapping\": {\n \"Estab\": \"establishment\",\n \"Insp #\": \"inspection_id\",\n \"POINTS\": \"score\",\n \"Inspector\": \"inspector\",\n \"INSPECTION DATE\": \"inspection_date\",\n \"Determination\": \"result\"\n },\n \"missing\": [\n \"critical_viol... |
schema_map-52eaee8700b50502 | <|start|><|user|>
Read the form and populate the fleet vehicle inventory schema as JSON. Do not reformat dates or amounts. Absent fields are null; values are copied as text exactly as written.
---
Target schema (fleet vehicle inventory):
unit_number, model_year, make, odometer, department, purchase_price, in_service
... | form_map | stable | deterministic | schema|20261019|3499 | 466 | 227 | 8b851501ae3d542e9290a53de06788775a1beee3b14297d376e9caeaa72b1918 | {"answer_form": "json", "frame_id": "form_map", "gold": "{\n \"unit_number\": \"U8286\",\n \"model_year\": \"2002\",\n \"make\": \"Ford\",\n \"odometer\": \"234917\",\n \"department\": \"Parks\",\n \"purchase_price\": \"$94,709.00\",\n \"in_service\": null\n}", "marker": "wrap:1", "n_cols": 6, "n_extra": 0, "n_missing"... |
schema_map-18365986cb436137 | <|start|><|user|>
Re-map the columns of this file onto the target schema and output the resulting CSV (header = schema fields in order). Values stay as they are. Leave missing fields empty and drop columns the schema does not define.
Excerpt:
Target schema (building permit):
{
"permit_number": "string (identifier)",
... | remap_csv | stable | deterministic | schema|20261019|21228 | 1,013 | 537 | 057844178907047f149da0bb30cd872e1e7229658a78d21ce217e8321c749315 | {"answer_form": "csv", "frame_id": "remap_csv", "gold": "permit_number,issue_date,site_address,applicant,contractor,work_type,valuation,fee_paid\nBP976166,1999-07-09,,Horace Kittredge,Bluefen Logistics Corp.,Addition,2143105.00,\nBP703370,2003-03-25,,Kendra Jessup,Driftwood Capital LP,Alteration,2397791.00,\nBP151590,2... |
schema_map-9f8454204d5a37f2 | Query: Produce the target-schema version of this CSV: rename and reorder the columns to match the schema, keep every value as written, add empty columns for missing fields, and omit columns the schema does not define. Leave missing fields empty and drop columns the schema does not define.
Passage:
Target schema (Feder... | remap_csv | stable | deterministic | schema|20261019|1712 | 837 | 306 | f0ace6e46d6eb0b25d77903eb3561973934b7256b62766e1e53a63766cb942f1 | {"answer_form": "csv", "frame_id": "remap_csv", "gold": "document_number,agency,document_type,publication_date,comment_deadline,page_count\n2019-59002,Forest Service,,,\"August 30, 2021\",\n2019-34202,Forest Service,,,\"March 25, 2010\",", "marker": "wrap:3|Query:", "n_cols": 5, "n_extra": 2, "n_missing": 3, "presentat... |
schema_map-fd101103e4362022 | Instruction: Map the columns of this CSV extract to the target schema below. Return JSON with "mapping" (each source column -> target field, or null if it has no target) and "missing" (target fields with no source column). Columns that match no target field map to null.
<text>
Target schema (government contract award)... | map_json | stable | deterministic | schema|20261019|15296 | 872 | 395 | d6a7590bc952bd627937e4f00bb037ecd4685cfd845fa1212ccf6c943febc4fe | {"answer_form": "json", "frame_id": "map_json", "gold": "{\"mapping\": {\"CONTRACT NO.\": \"contract_id\", \"dollars_obligated\": \"amount_usd\", \"Seatside\": \"set_aside\", \"PoP End\": \"end_date\", \"supplier\": \"vendor\", \"begin_date\": \"start_date\"}, \"missing\": [\"award_date\", \"agency\"]}", "marker": "wra... |
schema_map-e3a5d730a9b16dd3 | <|start|><|user|>
Spot the odd column out in the CSV below (name it, then one sentence of reasoning).
```
applicant_name,ISSUE DT,VALUATION,WK TYPE,Job Site Addr,Builder,Permit Num,ISBN,FEE PAID
Celia Kittredge,04-Oct-2017,"$1,516,304.00",Addition,"9952 Mill Ave, Brindle Point",Fenwick Grain Co.,BP259939,978-172573891... | odd_column | stable | deterministic | schema|20261019|12901 | 650 | 356 | 9add46e45d87141021fad6b517b13593b0e4d33647c474d3f41bcd2346998862 | {"answer_form": "prose", "frame_id": "odd_column", "gold": "\"ISBN\". Every other column is a building permit attribute; \"ISBN\" comes from a library catalog record.", "marker": "wrap:6", "n_cols": 9, "n_extra": 1, "n_missing": 0, "presentation": "chat", "source": "synthetic-typed-record", "stratum": "schema:permit", ... |
schema_map-ef8552c52dfabb9b | Task: Map and reorder: give the CSV rewritten to the grant award schema (schema field order), copying each value exactly; fields with no source column are empty; unmatched source columns are left out. Leave missing fields empty and drop columns the schema does not define.
Excerpt:
Target schema (grant award):
{
"gran... | remap_csv | stable | deterministic | schema|20261019|21116 | 1,019 | 458 | c4241953149b31824bffa3afc5aff4b6bf321f4dc6b2fa35eb9369bd707366fc | {"answer_form": "csv", "frame_id": "remap_csv", "gold": "grant_number,recipient,program,fiscal_year,amount_awarded,project_director,city\nG-195627,Fenwick Grain Co.,Historic Preservation,2013,1899634.00,\"Kittredge, Horace\",Stillwater Bend\nG-241702,Granite Arch Partners,Workforce Training,2010,2248634.00,\"Calloway, ... |
schema_map-fe3be70030a7074d | [Task]
Which target field does each column hold? Use the sample rows to decide. Reply with JSON ("mapping" and "missing"); columns that fit no field map to null. Columns that match no target field map to null.
Passage:
Target schema (food-safety inspection):
- inspection_id (string (identifier))
- establishment (strin... | map_json | stable | deterministic | schema|20261019|10925 | 755 | 317 | d94fd0160d2b013bb5ac56b17562d736e98b52e46535b86eeabf798ac08c17f8 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\"mapping\": {\"critical_violations\": \"critical_violations\", \"Inspecotr\": \"inspector\", \"Establishment\": \"establishment\", \"Visit Date\": \"inspection_date\", \"Totla Pts\": \"score\", \"Row ID\": null}, \"missing\": [\"inspection_id\", \"result\"]}", ... |
schema_map-eef2a3ed8de6beab | Which header doesn't fit this data set? Name it and give the reason in one sentence.
"""
AMOUNT AWARDED,Fsic Yr,Recipient Organizaiton,Award ID,RECIPIENT CITY,PROGRAM NAME,Blood Type,Proj iDr
1292641.00,2021,Juniper Ridge Energy,G-125107,Copper Mesa,Workforce Training,AB+,"Pembrook, Quinn"
2841635.00,2018,"Corvane Sys... | odd_column | stable | deterministic | schema|20261019|5826 | 609 | 299 | 4027633fd7dc9d89552ccefb023796ec300fc898ef4889f570255db59dcf1e2c | {"answer_form": "prose", "frame_id": "odd_column", "gold": "The odd column is \"Blood Type\" — medical data has nothing to do with a grant award.", "marker": "wrap:0", "n_cols": 8, "n_extra": 1, "n_missing": 0, "presentation": "plain", "source": "synthetic-typed-record", "stratum": "schema:grant", "typo": true} |
schema_map-385aeed24a7a5003 | <|start|><|user|>
Convert these labeled form entries into a JSON record that follows the schema below (null for fields the form lacks; extra form fields are dropped). Absent fields are null; values are copied as text exactly as written.
Text:
Target schema (government contract award):
contract_id, vendor, award_date, ... | form_map | stable | deterministic | schema|20261019|1065 | 552 | 273 | dc30d6beac9df0b9d6895abc9950fce5a9e3eb16737592249cc8c1f643980f55 | {"answer_form": "json", "frame_id": "form_map", "gold": "{\n \"contract_id\": \"GS-7638633\",\n \"vendor\": null,\n \"award_date\": null,\n \"start_date\": \"2/5/99\",\n \"end_date\": \"3/12/99\",\n \"amount_usd\": \"3411366.00\",\n \"agency\": \"Dept. of the Interior\",\n \"set_aside\": \"Yes\"\n}", "marker": "wrap:2"... |
schema_map-6fabc275124900ae | Find the column that is out of place in this table and explain briefly.
Text:
award_fy,Tide Height (ft),recipient_organization,Recipient City,Grnt Num,PROJECT DIRECTOR
2025,242,Hollowell Medical Group,Ferndale Hollow,G-807399,"Lockhart, Verna"
2021,3262,Fenwick Grain Co.,Harlan Ridge,G-368873,"Pembrook, Greta"
The od... | odd_column | stable | deterministic | schema|20261019|8784 | 314 | 156 | 6eae00ecc9e1829c8caca676f4382d08451ba51f7a55655bbbb69772b3b5025a | {"answer_form": "prose", "frame_id": "odd_column", "gold": "The odd column is \"Tide Height (ft)\" — tide table data has nothing to do with a grant award.", "marker": "wrap:2", "n_cols": 6, "n_extra": 1, "n_missing": 2, "presentation": "plain", "source": "synthetic-typed-record", "stratum": "schema:grant", "typo": fals... |
schema_map-69c83539b60eb280 | <|start|><|user|>
Normalize the headers of this Federal Register document index file to the schema below and say which fields are absent. Return {"mapping": {source_header: target_field_or_null}, "missing": [target_field, ...]}. Columns that match no target field map to null.
"""
Target schema (Federal Register docume... | map_json | stable | deterministic | schema|20261019|20096 | 786 | 374 | 86b45c475a47038b95e6e23bb29a2f049bbf93478908830c6e189c0744139bce | {"answer_form": "json", "frame_id": "map_json", "gold": "{\"mapping\": {\"DOC #\": \"document_number\", \"Row ID\": null, \"Comments\": null, \"Pg Ct\": \"page_count\", \"Pub Date\": \"publication_date\", \"Type\": \"document_type\", \"issuing_agency\": \"agency\", \"Cmt Deadline\": \"comment_deadline\"}, \"missing\": ... |
schema_map-3d49794fba119367 | The header row below is messy (abbreviations, typos, different order). Match each source column to a field of the target schema and report which target fields are not present. Answer as JSON: {"mapping": {...}, "missing": [...]}. Columns that match no target field map to null.
"""
Target schema (fleet vehicle inventor... | map_json | stable | deterministic | schema|20261019|14419 | 607 | 334 | b4b042540776cebf9046c215624a6f7ece39577ade5958baa084b5f4c882c676 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\n \"mapping\": {\n \"yr_model\": \"model_year\",\n \"service_status_flag\": \"in_service\",\n \"Batch\": null,\n \"ASSIGNED DEPT\": \"department\",\n \"PricePaid\": \"purchase_price\",\n \"VEH UNIT\": \"unit_number\",\n \"Odometer\": \"odometer\",\n \"M... |
schema_map-53774f40601441e7 | <|start|><|user|>
Read the form and populate the court case record schema as JSON. Do not reformat dates or amounts. Absent fields are null; values are copied as text exactly as written.
"""
Target schema (court case record):
- case_number (string (identifier))
- court (string (one of: Superior Court, District Court, ... | form_map | stable | deterministic | schema|20261019|17131 | 813 | 324 | 681a2a571da8b188f54941c5d0f03986bc4226a38b0dfc7401f522ca0b2fc2fb | {"answer_form": "json", "frame_id": "form_map", "gold": "{\"case_number\": \"CV-242071\", \"court\": null, \"filing_date\": \"April 10, 2021\", \"judge\": \"Greta Dunmore\", \"plaintiff\": \"Fenwick Grain Co.\", \"defendant\": \"Alma Kittredge\", \"case_type\": \"Tort\", \"status\": \"Stayed\"}", "marker": "wrap:0", "n... |
schema_map-e0aacb43abfeb480 | <|start|><|system|>
Answer accurately and concisely.<|eot|>
<|user|>
One column in this Federal Register document index extract does not belong with the others. Which one is it, and why? Answer in one sentence that names the column.
Passage:
Publication Daet,FR Dco No,Issuing Agency,PG CT,Wind Speed (mph)
2010-09-11,2... | odd_column | stable | deterministic | schema|20261019|9535 | 574 | 269 | cce330c4689ed7d6818dfeff6888347105e384f42d95d66cd56d7e516aaf62bc | {"answer_form": "prose", "frame_id": "odd_column", "gold": "The odd column is \"Wind Speed (mph)\" — weather observation data has nothing to do with a Federal Register document index.", "marker": "wrap:3", "n_cols": 5, "n_extra": 1, "n_missing": 2, "presentation": "chat", "source": "synthetic-typed-record", "stratum": ... |
schema_map-fa5e34124c9c3934 | <|start|><|user|>
Align these source columns with the government contract award schema. Output a JSON object whose "mapping" keys are the source headers exactly as written, plus a "missing" list. Columns that match no target field map to null.
"""
Target schema (government contract award):
contract_id, vendor, award_d... | map_json | stable | deterministic | schema|20261019|8610 | 808 | 409 | 9e2d56ad947361b52140ecc1e1349b3987c3b925498d1c3e339b96f79fedab6b | {"answer_form": "json", "frame_id": "map_json", "gold": "{\n \"mapping\": {\n \"Last Modified\": null,\n \"VndrNm\": \"vendor\",\n \"Set Aside?\": \"set_aside\",\n \"pop_end\": \"end_date\",\n \"PoP Start\": \"start_date\",\n \"Agency\": \"agency\"\n },\n \"missing\": [\n \"contract_id\",\n \"award_date\",\n \... |
schema_map-9c326b4afc48284b | Which target field does each column hold? Use the sample rows to decide. Reply with JSON ("mapping" and "missing"); columns that fit no field map to null. Columns that match no target field map to null.
Passage:
Target schema (government contract award):
contract_id, vendor, award_date, start_date, end_date, amount_us... | map_json | stable | deterministic | schema|20261019|17047 | 643 | 324 | 37e20082c98a1406cd1236606d83011351f3d8007e89da64e488c5a50f633d78 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\"mapping\": {\"Awardee aNme\": \"vendor\", \"Value USD\": \"amount_usd\", \"Award Date\": \"award_date\", \"PoP End\": \"end_date\", \"Contract ID\": \"contract_id\", \"Begin Date\": \"start_date\", \"Setaside\": \"set_aside\"}, \"missing\": [\"agency\"]}", "ma... |
schema_map-70680b75486b3fe7 | Map the form fields below to the schema and output JSON {field: value}; values unchanged, null for absent fields; ignore form fields outside the schema. Absent fields are null; values are copied as text exactly as written.
Passage:
Target schema (Federal Register document index):
document_number, agency, document_type... | form_map | stable | deterministic | schema|20261019|16299 | 466 | 191 | 7469c3ae5beee79a011a2e29aaf397f6c64000759ea0e2c86f1eca952502b0c0 | {"answer_form": "json", "frame_id": "form_map", "gold": "{\"document_number\": \"2019-90110\", \"agency\": \"Coast Guard\", \"document_type\": null, \"publication_date\": null, \"comment_deadline\": \"2022-04-01\", \"page_count\": null}", "marker": "wrap:3", "n_cols": 4, "n_extra": 1, "n_missing": 3, "presentation": "p... |
schema_map-3165176596b25f1f | Which target field does each column hold? Use the sample rows to decide. Reply with JSON ("mapping" and "missing"); columns that fit no field map to null. Columns that match no target field map to null.
Passage:
Target schema (fleet vehicle inventory):
{
"unit_number": "string (identifier)",
"model_year": "integer (... | map_json | stable | deterministic | schema|20261019|609 | 820 | 402 | 779dff98485dd59ea962c80eb37bc5c2ebcc309c5a42fcdd53905d6c6e5407ff | {"answer_form": "json", "frame_id": "map_json", "gold": "{\"mapping\": {\"PRICE PAID\": \"purchase_price\", \"Miles\": \"odometer\", \"Make\": \"make\", \"YR MODEL\": \"model_year\", \"active\": \"in_service\", \"UsingDepartment\": \"department\"}, \"missing\": [\"unit_number\"]}", "marker": "wrap:3", "n_cols": 6, "n_e... |
schema_map-e93ed67c0dae5324 | <|start|><|user|>
Read the form and populate the grant award schema as JSON. Do not reformat dates or amounts. Absent fields are null; values are copied as text exactly as written.
<text>
Target schema (grant award):
{
"grant_number": "string (identifier)",
"recipient": "string (organization)",
"program": "string (... | form_map | stable | deterministic | schema|20261019|16972 | 828 | 348 | 0beb113b60c7771a1f40fcb7a458329a60a178e70ff755e0353d1a6160bfaf97 | {"answer_form": "json", "frame_id": "form_map", "gold": "{\"grant_number\": \"G-298552\", \"recipient\": \"Fenwick Grain Co.\", \"program\": \"Historic Preservation\", \"fiscal_year\": \"2008\", \"amount_awarded\": \"$1,348,897.00\", \"project_director\": \"Dunmore, Kendra\", \"city\": \"Marlow Heights\"}", "marker": "... |
schema_map-47350ba653125e37 | Task: The header row below is messy (abbreviations, typos, different order). Match each source column to a field of the target schema and report which target fields are not present. Answer as JSON: {"mapping": {...}, "missing": [...]}. Columns that match no target field map to null.
```
Target schema (food-safety insp... | map_json | stable | deterministic | schema|20261019|16079 | 945 | 419 | b000da0c4565d1dae77fcda6a567747bb1c7ac76c306707309fe3abc2dac09f7 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\n \"mapping\": {\n \"Business Name\": \"establishment\",\n \"Outcome\": \"result\",\n \"Batch\": null,\n \"Notes\": null,\n \"inspector\": \"inspector\",\n \"Inspection Date\": \"inspection_date\",\n \"POINTS\": \"score\",\n \"# Critical\": \"critical_v... |
schema_map-8beff3474c55526c | The header row below is messy (abbreviations, typos, different order). Match each source column to a field of the target schema and report which target fields are not present. Answer as JSON: {"mapping": {...}, "missing": [...]}. Columns that match no target field map to null.
Passage:
Target schema (property tax asse... | map_json | stable | deterministic | schema|20261019|2046 | 962 | 532 | ff5186ec18bf21316163b4b3a08d1cd6babfde0d4f5bbeed192aa4fe786518bb | {"answer_form": "json", "frame_id": "map_json", "gold": "{\"mapping\": {\"Improvement Value\": \"improvement_value\", \"Ownr\": \"owner\", \"TAX EXEMPT FLAG\": \"exempt\", \"Yr\": \"tax_year\", \"Property Location\": \"situs_address\", \"Land\": \"land_value\", \"Prcl #\": \"parcel_id\"}, \"missing\": []}", "marker": "... |
schema_map-8c901c8f3377263e | <|start|><|user|>
Align these source columns with the fleet vehicle inventory schema. Output a JSON object whose "mapping" keys are the source headers exactly as written, plus a "missing" list. Columns that match no target field map to null.
Text:
Target schema (fleet vehicle inventory):
{
"unit_number": "string (ide... | map_json | stable | deterministic | schema|20261019|2516 | 1,049 | 483 | 37a29aee00fc90affb9382a2a6abbabf89c4ec70aeabb0a403b8741490b8ea29 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\"mapping\": {\"my\": \"model_year\", \"Acq Cost\": \"purchase_price\", \"FLEET NUMBER\": \"unit_number\", \"MILES\": \"odometer\", \"Mfr\": \"make\", \"Comments\": null, \"ASSIGNED DEPT\": \"department\", \"Notes\": null, \"In Service\": \"in_service\"}, \"miss... |
schema_map-4c8535dc733e1e67 | <|start|><|user|>
Map the columns of this CSV extract to the target schema below. Return JSON with "mapping" (each source column -> target field, or null if it has no target) and "missing" (target fields with no source column). Columns that match no target field map to null.
```
Target schema (Federal Register documen... | map_json | stable | deterministic | schema|20261019|3719 | 1,060 | 409 | 4884c7be8a9151259519c12e51651ecddf703e14bf03100e8b0ffa6a36944dc6 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\n \"mapping\": {\n \"FRDOC\": \"document_number\",\n \"Pbu Dt\": \"publication_date\",\n \"Agency Name\": \"agency\",\n \"Pages\": \"page_count\",\n \"comments_due\": \"comment_deadline\",\n \"Category\": \"document_type\"\n },\n \"missing\": []\n}", "mar... |
schema_map-db10888eb1c99643 | <|start|><|user|>
Map the form fields below to the schema and output JSON {field: value}; values unchanged, null for absent fields; ignore form fields outside the schema. Absent fields are null; values are copied as text exactly as written.
<text>
Target schema (property tax assessment):
{
"parcel_id": "string (ident... | form_map | stable | deterministic | schema|20261019|6728 | 702 | 316 | 3412c42e2fc51dcc0cde5e2f95c88387bbbdf7beaf22f89fbcd7889407674907 | {"answer_form": "json", "frame_id": "form_map", "gold": "{\"parcel_id\": \"P-96190327\", \"owner\": \"Marisol Abernathy\", \"situs_address\": null, \"tax_year\": \"1998\", \"land_value\": \"278771.00\", \"improvement_value\": \"1669428.00\", \"exempt\": \"TRUE\"}", "marker": "wrap:4", "n_cols": 6, "n_extra": 0, "n_miss... |
schema_map-e3b2be6bdc8c80c7 | <|start|><|user|>
Schema mapping task: for every column in the data, give the target field it corresponds to (null for columns the schema does not have), then list the schema fields the data lacks. JSON only. Columns that match no target field map to null.
Excerpt:
Target schema (court case record):
{
"case_number": ... | map_json | stable | deterministic | schema|20261019|22869 | 1,221 | 461 | 1ef0fd7961c0d9a013226791f0d4aa6b7436ded0a7b1a5ab4257cedce8cfdb45 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\"mapping\": {\"filed_on\": \"filing_date\", \"Defendnat\": \"defendant\", \"Case #\": \"case_number\", \"Plaintiff\": \"plaintiff\", \"Nature of Suit\": \"case_type\", \"STAT\": \"status\", \"Notes\": null}, \"missing\": [\"court\", \"judge\"]}", "marker": "wra... |
schema_map-755b47e0dc08e7b9 | Task: Extract the filled-in values of this form into JSON keyed by the target schema fields. Use null for schema fields the form does not include. Copy values exactly as written. Absent fields are null; values are copied as text exactly as written.
Text:
Target schema (grant award):
{
"grant_number": "string (identif... | form_map | stable | deterministic | schema|20261019|176 | 796 | 304 | ee41c608c3b610e2271d932f7d0542c807999b44a59d20ca535776f0c3a9f7e1 | {"answer_form": "json", "frame_id": "form_map", "gold": "{\"grant_number\": \"G-971105\", \"recipient\": \"Ironbark Freight Inc.\", \"program\": \"Broadband Access\", \"fiscal_year\": \"2018\", \"amount_awarded\": null, \"project_director\": \"Lockhart, Nolan\", \"city\": \"Ashby Crossing\"}", "marker": "wrap:2|Task:",... |
schema_map-e88da33534c30b44 | Question: Spot the odd column out in the CSV below (name it, then one sentence of reasoning).
```
Tide Height (ft),AcqCost,Service Status Flag,UNIT NO,ODO READING,MY,Dept,Mfr
3312,73279.00,No,U9228,197816,2002,Public Works,Ford
104,44051.00,Yes,U5336,50197,1997,Public Works,Chevrolet
713,53482.00,Yes,U3496,85902,2009,... | odd_column | stable | deterministic | schema|20261019|16386 | 453 | 302 | 774e3bba2c2056560f3751389918455d2b162ecc4a442a9bd4717ff8b92d0c05 | {"answer_form": "prose", "frame_id": "odd_column", "gold": "\"Tide Height (ft)\". Every other column is a fleet vehicle inventory attribute; \"Tide Height (ft)\" comes from a tide table record.", "marker": "wrap:6|Question:", "n_cols": 8, "n_extra": 1, "n_missing": 0, "presentation": "qa", "source": "synthetic-typed-re... |
schema_map-38f3c5613923e2d6 | Which of these columns would you flag as not belonging to a fleet vehicle inventory? Give the column name and a short reason.
Excerpt:
unit_num,assigned_dept,miles,MY,Tide Height (ft)
U5049,Parks,54954,2014,3202
U5237,Public Works,62840,2022,798
U8658,Public Works,43991,2007,1098
The odd column is "Tide Height (ft)" ... | odd_column | stable | deterministic | schema|20261019|13894 | 283 | 155 | addb5f09a66ab957da178f80e9b8577a02195095da3886ae535809d6612609ea | {"answer_form": "prose", "frame_id": "odd_column", "gold": "The odd column is \"Tide Height (ft)\" — tide table data has nothing to do with a fleet vehicle inventory.", "marker": "wrap:5", "n_cols": 5, "n_extra": 1, "n_missing": 3, "presentation": "plain", "source": "synthetic-typed-record", "stratum": "schema:vehicle"... |
schema_map-b41cf0f21b56ee73 | <|start|><|user|>
Spot the odd column out in the CSV below (name it, then one sentence of reasoning).
```
FeesCollected,Wind Speed (mph),issue_dt,Contractor,Wk Type,valuation
2582.00,4134,10/2/05,Elmstead Holdings,Demolition,2184832.00
2647.00,2659,11/25/09,Juniper Ridge Energy,Demolition,2056184.00
```<|eot|>
<|assis... | odd_column | stable | deterministic | schema|20261019|10206 | 327 | 166 | 43f36598d2b8e38cbf5bf884e0e160ecf1fd53dba42a5799d2082645371f5bbc | {"answer_form": "prose", "frame_id": "odd_column", "gold": "The odd column is \"Wind Speed (mph)\" — weather observation data has nothing to do with a building permit.", "marker": "wrap:6", "n_cols": 6, "n_extra": 1, "n_missing": 3, "presentation": "chat", "source": "synthetic-typed-record", "stratum": "schema:permit",... |
schema_map-8bc55256c96258a6 | <|start|><|user|>
Extract the filled-in values of this form into JSON keyed by the target schema fields. Use null for schema fields the form does not include. Copy values exactly as written. Absent fields are null; values are copied as text exactly as written.
Passage:
Target schema (government contract award):
contra... | form_map | stable | deterministic | schema|20261019|9353 | 569 | 263 | e360e335ab1bc035b8384adfd2ac3d39d9a52d0a9614ec98c90042d5fafb98d0 | {"answer_form": "json", "frame_id": "form_map", "gold": "{\n \"contract_id\": \"GS-2501903\",\n \"vendor\": null,\n \"award_date\": \"November 17, 2009\",\n \"start_date\": null,\n \"end_date\": null,\n \"amount_usd\": \"6092683.00\",\n \"agency\": \"General Services Administration\",\n \"set_aside\": \"Yes\"\n}", "mar... |
schema_map-cbc55935c6cf5fdd | [Task]
Map and reorder: give the CSV rewritten to the building permit schema (schema field order), copying each value exactly; fields with no source column are empty; unmatched source columns are left out. Fields the source lacks stay empty, and source columns outside the schema are dropped.
Text:
Target schema (build... | remap_csv | stable | deterministic | schema|20261019|16543 | 893 | 631 | 45486584dccc1d410c5568c809a24172c4d1bfdd98719189a1079e6b37c9ce18 | {"answer_form": "csv", "frame_id": "remap_csv", "gold": "permit_number,issue_date,site_address,applicant,contractor,work_type,valuation,fee_paid\nBP867115,23-Jul-2025,\"1074 Quarry Rd, Cobble Creek\",,,Demolition,,$698.00\nBP619375,18-Mar-2019,\"294 Commerce Blvd, Stillwater Bend\",,,Alteration,,\"$1,311.00\"\nBP584369... |
schema_map-ed84d6163c5cacc8 | <|start|><|user|>
Map the columns of this CSV extract to the target schema below. Return JSON with "mapping" (each source column -> target field, or null if it has no target) and "missing" (target fields with no source column). Columns that match no target field map to null.
Passage:
Target schema (food-safety inspect... | map_json | stable | deterministic | schema|20261019|16886 | 1,097 | 461 | 5f5200f2ce051a10bb43e06368537f5aef1f56fc6737d2e85ec74ccd8908b09f | {"answer_form": "json", "frame_id": "map_json", "gold": "{\"mapping\": {\"FACILITY NAME\": \"establishment\", \"InspectedBy\": \"inspector\", \"OUTCOME\": \"result\", \"Inspection Number\": \"inspection_id\", \"Inspection Score\": \"score\", \"critical_violations\": \"critical_violations\", \"INSPECTION DATE\": \"inspe... |
schema_map-18b8c77232ec131d | Map the columns of this CSV extract to the target schema below. Return JSON with "mapping" (each source column -> target field, or null if it has no target) and "missing" (target fields with no source column). Columns that match no target field map to null.
Passage:
Target schema (food-safety inspection):
inspection_i... | map_json | stable | deterministic | schema|20261019|21905 | 696 | 328 | feff5e981e466cf6fd5aff3bf418ef25be80e20eb3a9521eeef35c5b1cb0c969 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\"mapping\": {\"Establishment\": \"establishment\", \"Visit Date\": \"inspection_date\", \"INSP REF\": \"inspection_id\", \"Comments\": null, \"DETERMINATION\": \"result\", \"# CRITICAL\": \"critical_violations\", \"Inspector\": \"inspector\", \"Row ID\": null, ... |
schema_map-cb6ab7c732cc249b | <|start|><|system|>
You are a helpful assistant.<|eot|>
<|user|>
Align these source columns with the court case record schema. Output a JSON object whose "mapping" keys are the source headers exactly as written, plus a "missing" list. Columns that match no target field map to null.
```
Target schema (court case record... | map_json | stable | deterministic | schema|20261019|17142 | 1,127 | 462 | 5cfb6a2cec9653a12eb0553f14d3cfacb54b629af2d80527d880514cfeb8ae11 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\n \"mapping\": {\n \"DOCKET NUMBER\": \"case_number\",\n \"JUDGE\": \"judge\",\n \"respondent_defendant\": \"defendant\",\n \"file_date\": \"filing_date\",\n \"STATUS\": \"status\",\n \"nature_of_suit\": \"case_type\",\n \"Pl Name\": \"plaintiff\"\n },\n... |
schema_map-cca6abd3327873e3 | <|start|><|system|>
Answer accurately and concisely.<|eot|>
<|user|>
Which target field does each column hold? Use the sample rows to decide. Reply with JSON ("mapping" and "missing"); columns that fit no field map to null. Columns that match no target field map to null.
---
Target schema (building permit):
{
"permit... | map_json | stable | deterministic | schema|20261019|17004 | 1,070 | 464 | 25d8092e1a9bf7dbaee6008c4c739571df481993f57c2772df8d213820aeb910 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\"mapping\": {\"Locatoin\": \"site_address\", \"issue_dt\": \"issue_date\", \"permit_no\": \"permit_number\", \"Contractor\": \"contractor\", \"Requested By\": \"applicant\", \"FEE PAID\": \"fee_paid\", \"Scope\": \"work_type\"}, \"missing\": [\"valuation\"]}", ... |
schema_map-e0f7735391b55c9e | Which of these columns would you flag as not belonging to a court case record? Give the column name and a short reason.
Text:
Cs Type,court,Petitioner/Plaintiff,DEF,FiledOn,CaseStatus,ISBN
Tort,Circuit Court,Halvorsen Industrial LLC,Greta Abernathy,"May 13, 2025",Open,978-964803217
Small Claims,Superior Court,Halvorse... | odd_column | stable | deterministic | schema|20261019|1734 | 396 | 171 | 7c0049cc721e5e4b583944404813e072912f6da16c13985f81757167166aa923 | {"answer_form": "prose", "frame_id": "odd_column", "gold": "\"ISBN\". Every other column is a court case record attribute; \"ISBN\" comes from a library catalog record.", "marker": "wrap:2", "n_cols": 7, "n_extra": 1, "n_missing": 2, "presentation": "plain", "source": "synthetic-typed-record", "stratum": "schema:court_... |
schema_map-05a2c2a306238fc7 | Normalize the headers of this fleet vehicle inventory file to the schema below and say which fields are absent. Return {"mapping": {source_header: target_field_or_null}, "missing": [target_field, ...]}. Columns that match no target field map to null.
```
Target schema (fleet vehicle inventory):
unit_number, model_year... | map_json | stable | deterministic | schema|20261019|703 | 510 | 248 | 80213f055db0fd4bf3b397efb5ea1f556e6abf4ac304ecf9c8787b936e6d202f | {"answer_form": "json", "frame_id": "map_json", "gold": "{\"mapping\": {\"fleet_number\": \"unit_number\", \"Dept\": \"department\", \"MY\": \"model_year\", \"Odo Reading\": \"odometer\", \"SERVICE STATUS FLAG\": \"in_service\"}, \"missing\": [\"make\", \"purchase_price\"]}", "marker": "wrap:6", "n_cols": 5, "n_extra":... |
schema_map-b1bf5cfb83a077c7 | <|start|><|user|>
Re-map the columns of this file onto the target schema and output the resulting CSV (header = schema fields in order). Values stay as they are. Fields the source lacks stay empty, and source columns outside the schema are dropped.
"""
Target schema (fleet vehicle inventory):
{
"unit_number": "string... | remap_csv | stable | deterministic | schema|20261019|22839 | 985 | 634 | 69f6a7109decd0568779d44b21abdfc3ec0140cf77f25ffd32a5fd0663a48ffd | {"answer_form": "csv", "frame_id": "remap_csv", "gold": "unit_number,model_year,make,odometer,department,purchase_price,in_service\nU1353,1996,Chevrolet,78202,Parks,\"$35,316.00\",\nU6324,2005,Ram,23998,Parks,\"$44,690.00\",\nU2162,2022,Toyota,202476,Public Works,\"$37,470.00\",\nU3984,2009,Ford,76247,Police,\"$25,280.... |
schema_map-ebf8ec734b5a61c0 | [Task]
Map the columns of this CSV extract to the target schema below. Return JSON with "mapping" (each source column -> target field, or null if it has no target) and "missing" (target fields with no source column). Columns that match no target field map to null.
Passage:
Target schema (government contract award):
- ... | map_json | stable | deterministic | schema|20261019|11267 | 808 | 363 | 4ccbfdc8779b7b22c2205b2b953cfc31dca3cc9080c436836a1990bd98adc2f4 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\"mapping\": {\"AWARD DATE\": \"award_date\", \"PoP End\": \"end_date\", \"Small Business Set-Aside\": \"set_aside\", \"POP START\": \"start_date\", \"Value USD\": \"amount_usd\"}, \"missing\": [\"contract_id\", \"vendor\", \"agency\"]}", "marker": "wrap:3|[Task... |
schema_map-69d355ee2fd9eec5 | <|start|><|system|>
You check your work before answering.<|eot|>
<|user|>
Normalize the headers of this government contract award file to the schema below and say which fields are absent. Return {"mapping": {source_header: target_field_or_null}, "missing": [target_field, ...]}. Columns that match no target field map to... | map_json | stable | deterministic | schema|20261019|14717 | 909 | 477 | 943600a212f2a7cb598ea67c66890b50fec96d75652c083e9523da3f63def61e | {"answer_form": "json", "frame_id": "map_json", "gold": "{\n \"mapping\": {\n \"Complteion Date\": \"end_date\",\n \"Agency\": \"agency\",\n \"SB Set-Aside Flag\": \"set_aside\",\n \"award_date\": \"award_date\",\n \"VENDOR\": \"vendor\",\n \"Start Dt\": \"start_date\",\n \"obligated_amount\": \"amount_usd\",\n ... |
schema_map-8f9bd8e7764431b0 | <|start|><|system|>
You are a helpful assistant.<|eot|>
<|user|>
Map the columns of this CSV extract to the target schema below. Return JSON with "mapping" (each source column -> target field, or null if it has no target) and "missing" (target fields with no source column). Columns that match no target field map to nul... | map_json | stable | deterministic | schema|20261019|1551 | 1,124 | 440 | 05e4e7923a5ec6609f65d241c0e6ca37c6a3ca9f874ac327a86b875ccf303544 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\n \"mapping\": {\n \"Amount Awarded\": \"amount_awarded\",\n \"Notes\": null,\n \"Proj Dir\": \"project_director\",\n \"PROGRAM NAME\": \"program\",\n \"Award FY\": \"fiscal_year\",\n \"Grantee\": \"recipient\"\n },\n \"missing\": [\n \"grant_number\",\n... |
schema_map-fde7cd56208c0496 | <|start|><|user|>
Schema mapping task: for every column in the data, give the target field it corresponds to (null for columns the schema does not have), then list the schema fields the data lacks. JSON only. Columns that match no target field map to null.
---
Target schema (fleet vehicle inventory):
- unit_number (st... | map_json | stable | deterministic | schema|20261019|1383 | 755 | 324 | 26ff19ac3461d26914e1c05124cec942dcf3e349bb9fcb115743fae429da1b54 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\"mapping\": {\"Department\": \"department\", \"Miles\": \"odometer\", \"PricePaid\": \"purchase_price\", \"YR MODEL\": \"model_year\", \"Row ID\": null, \"ServiceStatusFlag\": \"in_service\"}, \"missing\": [\"unit_number\", \"make\"]}", "marker": "wrap:1", "n_c... |
schema_map-f7a4adf72c80fbe4 | Question: This file mixes in a column from a different kind of record. Identify it.
<text>
Judge,Dft aNme,ISBN,Date Filed,CASE TYPE,case_num,Petitioner / Plaintiff,TRIBUNAL,STAT
"Fairbanks, Celia","Ivers, Tamsin",978-196646491,"August 24, 2004",Tort,CV-818130,Granite Arch Partners,District Court,Stayed
"Ivers, Nolan",... | odd_column | stable | deterministic | schema|20261019|19208 | 448 | 212 | b1ca949a95772c7a87b59f6114741ecfb4ad1db227f9d776d7ea258231f99714 | {"answer_form": "prose", "frame_id": "odd_column", "gold": "The odd column is \"ISBN\" — library catalog data has nothing to do with a court case record.", "marker": "wrap:4|Question:", "n_cols": 9, "n_extra": 1, "n_missing": 0, "presentation": "qa", "source": "synthetic-typed-record", "stratum": "schema:court_case", "... |
schema_map-ea604eb9874d9dc3 | <|start|><|user|>
Re-map the columns of this file onto the target schema and output the resulting CSV (header = schema fields in order). Values stay as they are. Leave missing fields empty and drop columns the schema does not define.
---
Target schema (government contract award):
{
"contract_id": "string (identifier)... | remap_csv | stable | deterministic | schema|20261019|14393 | 1,232 | 748 | 8858af4b4d093fc10c3cd7db15cf6f93588b989910fe608886a8650ba6e1ad6c | {"answer_form": "csv", "frame_id": "remap_csv", "gold": "contract_id,vendor,award_date,start_date,end_date,amount_usd,agency,set_aside\nGS-3923254,Ironbark Freight Inc.,,\"August 4, 2000\",\"January 22, 2001\",8280944.00,Dept. of the Interior,\nGS-7568077,Granite Arch Partners,,\"March 18, 2016\",\"April 15, 2017\",751... |
schema_map-368e964e91eaa751 | <|start|><|user|>
Which target field does each column hold? Use the sample rows to decide. Reply with JSON ("mapping" and "missing"); columns that fit no field map to null. Columns that match no target field map to null.
```
Target schema (Federal Register document index):
- document_number (string (identifier))
- age... | map_json | stable | deterministic | schema|20261019|15886 | 936 | 391 | c2be8a97f7678aeb6fff7a2d83a56aa48bf500d4f45dbb44797a6c4a88528d65 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\n \"mapping\": {\n \"Pg Ct\": \"page_count\",\n \"Agency Name\": \"agency\",\n \"FR Doc No\": \"document_number\",\n \"ActionType\": \"document_type\",\n \"Pub Date\": \"publication_date\"\n },\n \"missing\": [\n \"comment_deadline\"\n ]\n}", "marker": "w... |
schema_map-f30fe7783e836077 | Map and reorder: give the CSV rewritten to the property tax assessment schema (schema field order), copying each value exactly; fields with no source column are empty; unmatched source columns are left out. Fields the source lacks stay empty, and source columns outside the schema are dropped.
Excerpt:
Target schema (p... | remap_csv | stable | deterministic | schema|20261019|8665 | 1,094 | 820 | 0110d94c99e06718adcf7f739f443ce39f89586a4ecf62d9cae4ae3751ff6137 | {"answer_form": "csv", "frame_id": "remap_csv", "gold": "parcel_id,owner,situs_address,tax_year,land_value,improvement_value,exempt\nP-87492410,\"Eastwick, Odette\",\"958 Quarry Rd, Larkspur\",1999,\"$328,344.00\",\"$121,454.00\",\nP-84429659,\"Redfern, Fletcher\",\"2306 Quarry Rd, Harlan Ridge\",2006,\"$510,108.00\",\... |
schema_map-215a6139b7629f11 | ### Task
Reshape this data to the schema below. Output CSV only, header first, with values unchanged. Leave missing fields empty and drop columns the schema does not define.
```
Target schema (Federal Register document index):
{
"document_number": "string (identifier)",
"agency": "string (one of: Environmental Prote... | remap_csv | stable | deterministic | schema|20261019|14111 | 680 | 244 | a1a6c61d144cf19758bb1277bb9df13ea07574826788d684f7abaca76d4e7ad8 | {"answer_form": "csv", "frame_id": "remap_csv", "gold": "document_number,agency,document_type,publication_date,comment_deadline,page_count\n,Forest Service,Proposed Rule,,08/19/2022,\n,Forest Service,Rule,,11/20/2021,", "marker": "wrap:6|### Task", "n_cols": 3, "n_extra": 0, "n_missing": 3, "presentation": "qa", "sourc... |
schema_map-c801fbb648e7e457 | Spot the odd column out in the CSV below (name it, then one sentence of reasoning).
```
Project Driector,FY,Recipient Organization,Award ID,Program,municipality,ISBN
Fletcher Eastwick,1995,Elmstead Holdings,G-618033,Workforce Training,Wexmoor,978-391755111
Bertram Abernathy,1991,Hollowell Medical Group,G-376393,Histor... | odd_column | stable | deterministic | schema|20261019|9336 | 649 | 326 | 1855f10e740f2a1e3f0c2462cb298a3f877d670fddab8288e64e09629b30493b | {"answer_form": "prose", "frame_id": "odd_column", "gold": "\"ISBN\". Every other column is a grant award attribute; \"ISBN\" comes from a library catalog record.", "marker": "wrap:6", "n_cols": 7, "n_extra": 1, "n_missing": 1, "presentation": "plain", "source": "synthetic-typed-record", "stratum": "schema:grant", "typ... |
schema_map-8d1a9d2fee935529 | Query: Produce the target-schema version of this CSV: rename and reorder the columns to match the schema, keep every value as written, add empty columns for missing fields, and omit columns the schema does not define. Leave missing fields empty and drop columns the schema does not define.
Passage:
Target schema (court... | remap_csv | stable | deterministic | schema|20261019|9901 | 1,312 | 697 | 5a308b59ac58226cbbf1701446dc1c7fc7b9b16ce74f573a3ddbb61ee36a2931 | {"answer_form": "csv", "frame_id": "remap_csv", "gold": "case_number,court,filing_date,judge,plaintiff,defendant,case_type,status\nCV-261141,,\"June 20, 2023\",Odette Lockhart,Elmstead Holdings,Edwina Redfern,Probate,Dismissed\nCV-688513,,\"June 29, 2019\",Lionel Abernathy,Hollowell Medical Group,Fletcher Pembrook,Prob... |
schema_map-eef3fbbc641e2736 | Extract the filled-in values of this form into JSON keyed by the target schema fields. Use null for schema fields the form does not include. Copy values exactly as written. Absent fields are null; values are copied as text exactly as written.
---
Target schema (food-safety inspection):
{
"inspection_id": "string (ide... | form_map | stable | deterministic | schema|20261019|11300 | 700 | 270 | 9ba6ae071bb31137e47af5c796c132ef85fd7fa1efb0a727ac30809d476c8c4f | {"answer_form": "json", "frame_id": "form_map", "gold": "{\"inspection_id\": \"INS8925250\", \"establishment\": null, \"inspection_date\": null, \"inspector\": null, \"score\": \"59\", \"result\": \"Pass with conditions\", \"critical_violations\": \"5\"}", "marker": "wrap:1", "n_cols": 6, "n_extra": 2, "n_missing": 3, ... |
schema_map-c44810d0f6363894 | Schema mapping task: for every column in the data, give the target field it corresponds to (null for columns the schema does not have), then list the schema fields the data lacks. JSON only. Columns that match no target field map to null.
Excerpt:
Target schema (food-safety inspection):
- inspection_id (string (identi... | map_json | stable | deterministic | schema|20261019|4585 | 990 | 465 | d97cf02486fc4c5ba26f03f8c5bd1eeeeff1efd6adc7747993dcd621c49d4b97 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\"mapping\": {\"Entered By\": null, \"InspRef\": \"inspection_id\", \"VISIT DATE\": \"inspection_date\", \"Inspected By\": \"inspector\", \"Critical Violaitons\": \"critical_violations\", \"Row ID\": null, \"DETERMINATION\": \"result\"}, \"missing\": [\"establis... |
schema_map-c09d80915f2b3c7c | Task: Read the form and populate the grant award schema as JSON. Do not reformat dates or amounts. Absent fields are null; values are copied as text exactly as written.
Excerpt:
Target schema (grant award):
- grant_number (string (identifier))
- recipient (string (organization))
- program (string (one of: Rural Water,... | form_map | stable | deterministic | schema|20261019|19567 | 762 | 301 | 70aa79d7df612aa66aed50594911f4743d1345669af41413e64b0ded35a4896f | {"answer_form": "json", "frame_id": "form_map", "gold": "{\"grant_number\": \"G-996394\", \"recipient\": \"Bluefen Logistics Corp.\", \"program\": \"Workforce Training\", \"fiscal_year\": null, \"amount_awarded\": \"2617369.00\", \"project_director\": \"Horace Ivers\", \"city\": \"Ashby Crossing\"}", "marker": "wrap:5|... |
schema_map-d03cb03a3b649213 | Query: Which header doesn't fit this data set? Name it and give the reason in one sentence.
---
ISBN,Nature of Suit,court,Respondent / Defednant,DISPOSITION STATUS,DateFiled,Presiding Judge,Plaintfif,docket_number
978-849150154,Tort,Superior Court,"Merriweather, Marisol",Open,8/25/98,"Dunmore, Tamsin",Granite Arch Par... | odd_column | stable | deterministic | schema|20261019|4922 | 715 | 338 | bf18943bd3fd438d182168edb4781a927e59179b4bbc60c8f65c0480dd4971f3 | {"answer_form": "prose", "frame_id": "odd_column", "gold": "The odd column is \"ISBN\" — library catalog data has nothing to do with a court case record.", "marker": "wrap:1|Query:", "n_cols": 9, "n_extra": 1, "n_missing": 0, "presentation": "qa", "source": "synthetic-typed-record", "stratum": "schema:court_case", "typ... |
schema_map-5a43e576a60fe0f2 | Schema mapping task: for every column in the data, give the target field it corresponds to (null for columns the schema does not have), then list the schema fields the data lacks. JSON only. Columns that match no target field map to null.
---
Target schema (government contract award):
contract_id, vendor, award_date, ... | map_json | stable | deterministic | schema|20261019|17160 | 677 | 364 | 290aca04c5acbfc0e5b9cd1601c368de76bdefb0ed96967dd6dd1f8487f11bc3 | {"answer_form": "json", "frame_id": "map_json", "gold": "{\n \"mapping\": {\n \"Comments\": null,\n \"PoP End\": \"end_date\",\n \"vndr_nm\": \"vendor\",\n \"set_aside\": \"set_aside\",\n \"Award Number\": \"contract_id\",\n \"Row ID\": null,\n \"PoP Start\": \"start_date\",\n \"Obligated Amount\": \"amount_usd... |
schema_map-7ce3224d46aa9af3 | Task: Align these source columns with the court case record schema. Output a JSON object whose "mapping" keys are the source headers exactly as written, plus a "missing" list. Columns that match no target field map to null.
---
Target schema (court case record):
case_number, court, filing_date, judge, plaintiff, defen... | map_json | stable | deterministic | schema|20261019|13964 | 767 | 373 | fffbfdb5e70c724c61168e36d00e7efe03c7a0da0ee04e9226de3d5419aa511c | {"answer_form": "json", "frame_id": "map_json", "gold": "{\n \"mapping\": {\n \"court\": \"court\",\n \"Statsu\": \"status\",\n \"TYPE OF CASE\": \"case_type\",\n \"FiledOn\": \"filing_date\",\n \"PlName\": \"plaintiff\",\n \"case_num\": \"case_number\",\n \"RESPONDENT / DEFENDANT\": \"defendant\",\n \"jdg\": \... |
schema_map-0697ffb0a89c7f81 | <|start|><|user|>
Rewrite this CSV to the target schema: columns in schema order with the schema's field names, values copied unchanged, empty cells for fields the source lacks, and source columns outside the schema dropped. Leave missing fields empty and drop columns the schema does not define.
Text:
Target schema (p... | remap_csv | stable | deterministic | schema|20261019|4678 | 806 | 414 | c1be3db8ff8f9310b120f8e9ab8ea585aac53c247940b2245dc0bba514241f40 | {"answer_form": "csv", "frame_id": "remap_csv", "gold": "parcel_id,owner,situs_address,tax_year,land_value,improvement_value,exempt\nP-10713432,,\"7296 Ridge Ln, Harlan Ridge\",2025,662348.00,1445901.00,1\nP-48444815,,\"8206 Mill Ave, Cedar Gap\",1992,694433.00,917474.00,1", "marker": "wrap:2", "n_cols": 6, "n_extra": ... |
schema_map-7c208a10c72dc413 | [Task]
Find the column that is out of place in this table and explain briefly.
Passage:
Horsepower,Tax aMp No,Property Location,bldg_value,ASSESSMENT YEAR,Exmpt
1976,P-31822260,"8727 Harbor Way, Wexmoor","$998,296.00",2012,1
2220,P-72952522,"3769 Ridge Ln, Wexmoor","$427,693.00",2021,1
[Response]
The odd column is "H... | odd_column | stable | deterministic | schema|20261019|40 | 300 | 172 | 5f9c700f889fa9498b30a6be8bbf2e16a0cd7f191920354642003c573083d114 | {"answer_form": "prose", "frame_id": "odd_column", "gold": "The odd column is \"Horsepower\" — fleet vehicle data has nothing to do with a property tax assessment.", "marker": "wrap:3|[Task]", "n_cols": 6, "n_extra": 1, "n_missing": 2, "presentation": "qa", "source": "synthetic-typed-record", "stratum": "schema:propert... |
End of preview. Expand in Data Studio
ALEA schema mapping
Messy spreadsheet columns (case and separator variants, typos, reordered, extra and missing columns) are mapped to a target schema for eight fictional public-sector record types (court case, contract award, building permit, grant, food-safety inspection, property tax, Federal Register index, fleet vehicle). Tasks: column mapping as JSON with missing fields, CSV rewritten to the schema, form to schema JSON with nulls, and spotting a column from an unrelated record type.
22,755 documents, 8,816,405 tokens (alea-v2b-sw32k, the ALEA R4 tokenizer, counting each document's start and end tokens).
Part of the ALEA R4 training mix
This dataset contains exactly the documents of this family that ALEA's R4 model trained on in its stable pretraining mixture (second data stage, one pass each). Every row is byte-identical to the trained text; text_sha256 is the SHA-256 of text.
Held-out evaluation items generated by the same code are not part of this release.
Composition
| task | documents | tokens |
|---|---|---|
map_json |
8,458 | 3,550,764 |
remap_csv |
5,988 | 3,017,250 |
form_map |
4,179 | 1,230,645 |
odd_column |
4,130 | 1,017,746 |
8,069 documents use ALEA's chat template.
How it was made
Written by deterministic code from typed records; no language model was used. Each frame states its null / empty / drop convention.
Sources
- Invented typed records with hand-written field aliases
Verification
- Rewritten CSV answers are parsed back and compared with the typed records; aliases are unique within each schema.
- Decontamination: training rows that shared a word 13-gram with a held-out evaluation item of this or any sibling family were removed, and rows overlapping ALEA's evaluation sets were removed or logged according to ALEA's decontamination policy.
- Sampled records were reviewed blind by a separate reviewer who recomputed or re-derived the answers; defect patterns found in review were fixed in the generator, or the affected rows dropped, before the final build.
Columns
| column | content |
|---|---|
id |
row id from the generator (not always unique: a few families reuse an id across variants, and a handful of exact duplicate documents were trained twice and appear twice) |
text |
the document exactly as trained; chat documents use ALEA's chat template, whose control tokens appear as text (`< |
task |
task kind |
phase |
stable (trained in R4's stable pretraining mixture) |
generator |
what wrote the row (deterministic = code only) |
group_id |
source passage / record group; all variants of one group share it |
prompt_chars |
length in characters of the request part of text (0 for plain documents; null where not recorded) |
n_tokens |
tokens under alea-v2b-sw32k, including the document start and end tokens |
text_sha256 |
SHA-256 of text |
meta |
JSON: the generator's structured fields (source, presentation, answer form, gold values, checks) |
Known limitations
- The documents are generated and more repetitive than natural text.
License
CC-BY-4.0 for this compilation and the generated text. Public-domain and public-record source text keeps its status; Wikidata is CC0; OpenGloss and SHELF are CC-BY-4.0.
Published by the ALEA Institute.
- Downloads last month
- 15