The whole of this run is a release asset
This site holds 3 of the 31 question directories this run wrote; the archive holds every one:
archive
minidev-sqlite-gpt-4-turbo.tar.gz size
12,234,405 bytes
sha256
4f44c336f425afe237d3fc1cb12c528c960d4f4a373a5f55d8c580eb73af4c63
What it counted
what how many
questions audited 52 NOT_EQUAL 31 EQUAL 18 ERROR 3 probes fired 3
the verdicts of run audit-fb23ec3b-e8c5-4809-a8bc-a5db5787fee3, from summary.json 3 multiplicity 28 NOT_EQUAL 18 EQUAL 3 ERROR Of the questions this run audited: 3 multiplicity, 28 NOT_EQUAL, 18 EQUAL, 3 ERROR
the probes
probe fired on
ordering-over-numeric-text 0 arbitrary-cut 1 not-a-function-of-the-data 2 float-aggregate-order 0 direction-against-question 0
how many golds each probe fired on, of 52, from summary.json 0 ordering-over-numeric-text 1 arbitrary-cut 2 not-a-function-of-the-data 0 float-aggregate-order 0 direction-against-question Of 52 golds this run read, the probes fired on: arbitrary-cut on 1; not-a-function-of-the-data on 2.
credited by BIRD's own check and NOT_EQUAL here
what how many
credited by BIRD and NOT_EQUAL here 9 multiplicity 9 type 0 order 0 truncation 0 other 0 the test-suite check answered 1 0 the test-suite check answered 0 9
What the run was made of
run
audit-fb23ec3b-e8c5-4809-a8bc-a5db5787fee3
summary format
attestql/audit/summary/2
question file
data/questions/mini_dev_sqlite.json sha256:4ba5fa8de55856222f484d380d2ba872b380bf79d825de70478e2120cb0fc43b | https://bird-bench.oss-cn-beijing.aliyuncs.com/minidev.zip (sha256 cc48ba16838204e4e214512030cb572eeb5f7bcdd999bae4b9b6ff12ec13b92f, downloaded 2026-09-07), member minidev/MINIDEV/mini_dev_sqlite.json, 2024-06-19
questions in the file
500
prediction file
data/preds-sqlite/predict_mini_dev_gpt-4-turbo_sqlite.json sha256:ef9fcb0470d9710c8a9674fd35874ec0986925e27a4d5d9ccece8bff606fb5df | https://raw.githubusercontent.com/bird-bench/mini_dev/b3d4bcbbae9a96934ad812551eb400c7a3b23c12/llm/exp_result/sql_output_kg/predict_mini_dev_gpt-4-turbo_sqlite.json, 2024-06-19
statements read
500
keyed by
position
data file
data/minidev/dev_databases/card_games/card_games.sqlite sha256:c98bdb57fe7474da798b407785544b9af0daaad5d61fd21e2a73309493bc1227 | minidev.zip (sha256 cc48ba16838204e4e214512030cb572eeb5f7bcdd999bae4b9b6ff12ec13b92f), member minidev/MINIDEV/dev_databases, card_games/card_games.sqlite, 2024-06-13
server
SQLite 3.53.4 | file=/private/tmp/attestql-runs/data/minidev/dev_databases/card_games/card_games.sqlite | size=261820416
database role
file
engine
sqlite
parser
validator audit:sqlglot-sqlite-parse, sqlglot 30.18.0, dialect sqlite
serialization
attestql/audit/2
statement timeout
30 s
fixture digest depth
counts
schema digest
sha256:0654f2e002e823f47f4f1299e259b9557e98f1e388ac16be7a3fa11284be1030
data as of
2026-09-08T05:21:28.514364+00:00 the instant the run started
shuffled copies
prepared
shuffle seed
1
experimental probe
off
The session the run was made in
what the engine reported, 9 settings
sqlite_version
3.53.4
encoding
UTF-8
reverse_unordered_selects
0
query_only
1
journal_mode
wal
data_version
3
compile_options
ATOMIC_INTRINSICS=1,COMPILER=clang-21.0.0,DEFAULT_AUTOVACUUM,DEFAULT_CACHE_SIZE=-2000,DEFAULT_FILE_FORMAT=4,DEFAULT_JOURNAL_SIZE_LIMIT=-1,DEFAULT_MMAP_SIZE=0,DEFAULT_PAGE_SIZE=4096,DEFAULT_PCACHE_INITSZ=20,DEFAULT_RECURSIVE_TRIGGERS,DEFAULT_SECTOR_SIZE=4096,DEFAULT_SYNCHRONOUS=2,DEFAULT_WAL_AUTOCHECKPOINT=1000,DEFAULT_WAL_SYNCHRONOUS=2,DEFAULT_WORKER_THREADS=0,DIRECT_OVERFLOW_READ,ENABLE_API_ARMOR,ENABLE_COLUMN_METADATA,ENABLE_DBSTAT_VTAB,ENABLE_FTS3,ENABLE_FTS3_PARENTHESIS,ENABLE_FTS5,ENABLE_GEOPOLY,ENABLE_MATH_FUNCTIONS,ENABLE_MEMORY_MANAGEMENT,ENABLE_PERCENTILE,ENABLE_PREUPDATE_HOOK,ENABLE_RTREE,ENABLE_SESSION,ENABLE_STAT4,ENABLE_UNLOCK_NOTIFY,MALLOC_SOFT_LIMIT=1024,MAX_ATTACHED=10,MAX_COLUMN=2000,MAX_COMPOUND_SELECT=500,MAX_DEFAULT_PAGE_SIZE=8192,MAX_EXPR_DEPTH=1000,MAX_FUNCTION_ARG=1000,MAX_LENGTH=1000000000,MAX_LIKE_PATTERN_LENGTH=50000,MAX_MMAP_SIZE=0x7fff0000,MAX_PAGE_COUNT=0xfffffffe,MAX_PAGE_SIZE=65536,MAX_SQL_LENGTH=1000000000,MAX_TRIGGER_DEPTH=1000,MAX_VARIABLE_NUMBER=250000,MAX_VDBE_OP=250000000,MAX_WORKER_THREADS=8,MUTEX_PTHREADS,SYSTEM_MALLOC,TEMP_STORE=1,THREADSAFE=1,USE_URI
collation_list
RTRIM,NOCASE,BINARY
case_sensitive_like
0
What the run states about itself
ids the question file states twice
137, 138
prediction positions not compared
487, 488
stopped by the statement timeout, gold
518 the run's bound was 30 s
The questions
question
rule
verdict
class
probes
what was asked
q358 R-SET
NOT_EQUAL multiplicity What is the border color of card "Ancestor's Chosen"?
q440 R-SET
NOT_EQUAL multiplicity Which foreign language used by "A Pedra Fellwar"?
q462 R-SET
NOT_EQUAL multiplicity What's the Italian name of the set of cards with "Ancestor's Chosen" is in?
q371
ERROR prediction: statement: the text does not parse: Error tokenizing 'potlight = 1) AS percentage_french_story_spotligh'
q483
ERROR prediction: execute: execute: no such column: r.language
q518
ERROR gold: execute: execute: the statement ran past its 30s timeout
A question whose statement could not be run wrote no directory, so it is a row here and has no page: the summary holds the side that stopped, the step it stopped at and the engine's own message.
The same index, restricted
Each of these is a page of its own, so a filtered view has an address a reader can send. Nothing here is done by a script or by a query string.
The JSON this page was rendered from