Hugging Face's logo

Datasets:
HappyWindman555
/
d3-hub-evals

Tags:
Eval Results
decision-model
License:
Dataset card Data Studio Files
xet
 Community
Dataset Preview
Duplicate
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed
Error code:   DatasetGenerationError
Exception:    CastError
Message:      Couldn't cast
bench: string
model: string
n: int64
answered: int64
errors: int64
correct: int64
accuracy: double
latency_p50_s: double
results_sha256: string
probs: struct<A: double, B: double, C: double, D: double>
  child 0, A: double
  child 1, B: double
  child 2, C: double
  child 3, D: double
group: string
gold: string
pred: string
id: string
latency_s: double
to
{'id': Value('string'), 'gold': Value('string'), 'group': Value('string'), 'pred': Value('string'), 'probs': {'A': Value('float64'), 'B': Value('float64'), 'C': Value('float64'), 'D': Value('float64')}, 'latency_s': Value('float64')}
because column names don't match
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1827, in _prepare_split_single
                  for key, table in generator:
                                    ^^^^^^^^^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 613, in wrapped
                  for item in generator(*args, **kwargs):
                              ~~~~~~~~~^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 343, in _generate_tables
                  self._cast_table(pa_table, json_field_paths=json_field_paths),
                  ~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 132, in _cast_table
                  pa_table = table_cast(pa_table, self.info.features.arrow_schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2378, in table_cast
                  return cast_table_to_schema(table, schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2306, in cast_table_to_schema
                  raise CastError(
                  ...<3 lines>...
                  )
              datasets.table.CastError: Couldn't cast
              bench: string
              model: string
              n: int64
              answered: int64
              errors: int64
              correct: int64
              accuracy: double
              latency_p50_s: double
              results_sha256: string
              probs: struct<A: double, B: double, C: double, D: double>
                child 0, A: double
                child 1, B: double
                child 2, C: double
                child 3, D: double
              group: string
              gold: string
              pred: string
              id: string
              latency_s: double
              to
              {'id': Value('string'), 'gold': Value('string'), 'group': Value('string'), 'pred': Value('string'), 'probs': {'A': Value('float64'), 'B': Value('float64'), 'C': Value('float64'), 'D': Value('float64')}, 'latency_s': Value('float64')}
              because column names don't match
              
              The above exception was the direct cause of the following exception:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1369, in compute_config_parquet_and_info_response
                  parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
                                                                        ~~~~~~~~~~~~~~~~~~~~~~~~~^
                      builder, max_dataset_size_bytes=max_dataset_size_bytes
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  )
                  ^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 948, in stream_convert_to_parquet
                  builder._prepare_split(split_generator=splits_generators[split], file_format="parquet")
                  ~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1694, in _prepare_split
                  for job_id, done, content in self._prepare_split_single(
                                               ~~~~~~~~~~~~~~~~~~~~~~~~~~^
                      gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  ):
                  ^
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1879, in _prepare_split_single
                  raise DatasetGenerationError("An error occurred while generating the dataset") from e
              datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

id
string
gold
string
group
string
pred
string
probs
dict
latency_s
float64
55431e4f-6cc2-4218-b081-fc6e8e9a9547
B
de
D
{ "A": 0.1593805921736436, "B": 0.3522845499307601, "C": 0.11437844718249973, "D": 0.37395641071309654 }
1.423652
0caa89a2-7bf3-46d3-9567-fa3d4a7c4df4
C
de
A
{ "A": 0.58692094066902, "B": 0.21045453973581832, "C": 0.1441526314172986, "D": 0.05847188817786305 }
3.204503
49e4240f-f9a0-4edf-98b3-ab39e9c436e2
A
de
C
{ "A": 0.019965340998914505, "B": 0.35117996309703514, "C": 0.6238205753700504, "D": 0.005034120533999994 }
3.204263
61eca694-ca73-433a-aba1-c3da0fe04d1c
D
de
D
{ "A": 0.1675759364062928, "B": 0.2169609436801993, "C": 0.10878583434214267, "D": 0.5066772855713652 }
3.193626
3651f6dd-458c-4e85-8577-a4d383a0fe01
A
en
A
{ "A": 0.961435749575177, "B": 0.0011193254172048007, "C": 0.03641204551423572, "D": 0.001032879493382452 }
3.204317
ee9c501e-8d41-49d3-85bc-a4a1e8f6a57c
C
en
C
{ "A": 0.0021436598081283616, "B": 0.001832546738111235, "C": 0.9937928684654457, "D": 0.0022309249883146233 }
3.203875
08de014d-c169-4b01-be5b-9d4b49107c47
A
de
B
{ "A": 0.017750030655353233, "B": 0.8889581450247392, "C": 0.02623231941009785, "D": 0.06705950490980975 }
3.191991
8f3d2d74-6c8b-4eeb-8169-0f2a36cbb2c6
B
de
B
{ "A": 0.21989652799019316, "B": 0.5121980979169158, "C": 0.10140540799020664, "D": 0.16649996610268433 }
3.186209
430dc3d6-009b-4da6-8eba-a01f1c58589c
D
en
D
{ "A": 0.0030533734023105302, "B": 0.0013693858597449864, "C": 0.0037515144169765445, "D": 0.991825726320968 }
3.202725
d73c3f68-eda7-4e35-b048-6cf216e20e11
C
de
B
{ "A": 0.13270167587587742, "B": 0.3580246963154115, "C": 0.20785301641257178, "D": 0.3014206113961393 }
3.205588
7a856787-c785-4c10-8d02-803cc1211f02
D
de
A
{ "A": 0.5721480641506658, "B": 0.10466388567531432, "C": 0.15056672722801237, "D": 0.17262132294600743 }
3.188037
cfdc3e62-108c-4a28-b803-67377adfd0bc
B
en
B
{ "A": 0.058846526419875186, "B": 0.8526311260254136, "C": 0.007529776355660609, "D": 0.08099257119905064 }
3.186862
05ae153f-913a-45d9-a106-7d0f59df58d0
B
de
B
{ "A": 0.050852652974185154, "B": 0.7791103439199164, "C": 0.12334500887099195, "D": 0.0466919942349065 }
3.188061
521fda19-2122-4036-9ca2-548372b3f119
A
en
A
{ "A": 0.9718054686896406, "B": 0.0050907963247537975, "C": 0.01589120291721222, "D": 0.007212532068393345 }
3.200367
f1bdfe14-a0bc-496c-b9ea-49ed1d6d03df
B
en
B
{ "A": 0.02135193151646464, "B": 0.7695174531521873, "C": 0.19801800712688056, "D": 0.011112608204467512 }
3.205435
762d3adb-7523-49de-9b66-ec4826a8a3bb
D
en
D
{ "A": 0.14172319608347098, "B": 0.14279538335906145, "C": 0.3067173653136889, "D": 0.40876405524377857 }
3.195319
7e2f2168-4eeb-4b40-b1cc-bbc009a5a2ae
D
en
B
{ "A": 0.052137947172064644, "B": 0.6582892201424801, "C": 0.032455322423959786, "D": 0.2571175102614954 }
3.214228
ea8f2977-28a3-4e28-91f4-096f41844a19
A
en
A
{ "A": 0.7332573720976066, "B": 0.1860099728300912, "C": 0.009257164933395035, "D": 0.07147549013890718 }
3.206676
f4386097-9a9a-4ba7-b081-e0528333c7b4
D
en
D
{ "A": 0.09445294851570367, "B": 0.008957771515882463, "C": 0.04694875228620956, "D": 0.8496405276822042 }
3.196148
c60a0867-a71a-43e1-9bf9-4133628413ed
C
en
B
{ "A": 0.027697360398968227, "B": 0.5140318069473419, "C": 0.4056751279602007, "D": 0.05259570469348927 }
3.212233
b4ac3f55-14de-4683-8c14-ce2ccfbd17d5
C
de
B
{ "A": 0.28464697725180194, "B": 0.33616336253487206, "C": 0.27457899408038156, "D": 0.10461066613294437 }
4.696782
5171dc9e-c1ad-494a-9d96-1765550b65ac
D
en
D
{ "A": 0.4070423205228006, "B": 0.07924955645038999, "C": 0.005261729764413692, "D": 0.5084463932623957 }
4.697822
8817b2da-0807-4924-8fb1-82b0a8a68d3d
D
de
C
{ "A": 0.25325402894375454, "B": 0.047679007377784945, "C": 0.4287895474226907, "D": 0.2702774162557699 }
4.697387
c33c80ba-fd3b-45a0-9604-a488950ba5f0
C
de
A
{ "A": 0.45361831810025316, "B": 0.18828952662979137, "C": 0.2656153848145465, "D": 0.09247677045540895 }
4.696916
d600892c-cb99-4dee-a373-f31dff6b029f
D
en
D
{ "A": 0.026866369201803207, "B": 0.009966100873584136, "C": 0.024303984389181705, "D": 0.9388635455354308 }
4.697925
35b68326-6bba-4984-9ee5-accaff1ac71c
D
de
D
{ "A": 0.025099244732154677, "B": 0.030941924595301835, "C": 0.146417892068112, "D": 0.7975409386044315 }
3.304014
59d5682d-1406-42ae-9d74-a94a30670d4b
C
en
C
{ "A": 0.0205446995227143, "B": 0.02210515484276537, "C": 0.7002599512081209, "D": 0.2570901944263994 }
4.695243
bd2e156e-3b5d-4b89-9926-544bc28a2849
C
de
B
{ "A": 0.010282096786111717, "B": 0.9300420487083462, "C": 0.05750951312925078, "D": 0.0021663413762914585 }
4.69533
7086f62e-d79c-4a52-ae6e-ed7001daa738
A
en
A
{ "A": 0.955009438788072, "B": 0.004575202906121021, "C": 0.02249068426290679, "D": 0.017924674042900268 }
4.697041
5f67faea-52c7-43b3-b9d0-3e72ec742d4f
B
de
D
{ "A": 0.11663243308157706, "B": 0.18214679255492916, "C": 0.07461891692136087, "D": 0.6266018574421328 }
4.699605
ab120142-79f6-4cea-a940-fc483e65e098
C
de
A
{ "A": 0.5302334580266663, "B": 0.2678940295079439, "C": 0.12410961216484731, "D": 0.07776290030054249 }
4.698739
04999df8-83d2-45bc-abae-cb8e0308f96e
B
de
A
{ "A": 0.8937730357005723, "B": 0.04813340647846076, "C": 0.05416451986242061, "D": 0.0039290379585463445 }
4.699958
9975be1d-0c23-4dba-850f-b6e4319c3b4d
D
en
D
{ "A": 0.0937534943155193, "B": 0.0894279656341827, "C": 0.02206474836707815, "D": 0.7947537916832198 }
4.7034
b766bab3-7116-42ee-853e-2eb3f01c0c15
A
en
C
{ "A": 0.01650102561797309, "B": 0.010820274606555226, "C": 0.954831306156996, "D": 0.017847393618475607 }
2.955485
2cdd4813-3450-4626-8aca-773dfea1f607
B
en
B
{ "A": 0.03929122095150793, "B": 0.8749015991143398, "C": 0.012428112346183752, "D": 0.07337906758796858 }
2.954212
fc709281-f30d-4f31-9b2e-e02a643d00d6
A
de
D
{ "A": 0.12111095007748873, "B": 0.35560759056063557, "C": 0.1327873581091769, "D": 0.3904941012526988 }
2.956349
53bbecbc-ef12-4b60-9645-04015eeaa4ae
A
en
A
{ "A": 0.900343731851876, "B": 0.031296537628786626, "C": 0.01585433986217589, "D": 0.05250539065716153 }
2.949813
4424a6c0-5acb-4623-9b83-c9d345290889
A
de
C
{ "A": 0.1262759372249316, "B": 0.17815099509051463, "C": 0.37463846947926177, "D": 0.3209345982052919 }
2.953356
b9ad3761-ad5c-4fb5-8180-a2fcf31eadd2
B
de
B
{ "A": 0.0050049657616950015, "B": 0.8533088187879184, "C": 0.12592716751733796, "D": 0.015759047933048654 }
2.960395
03dffd6b-ba84-401d-85a0-2a66ce1929a1
C
en
C
{ "A": 0.002218759725792695, "B": 0.005622058323122066, "C": 0.9686367812817874, "D": 0.0235224006692979 }
2.952237
7e4940e6-2a92-4204-bef7-f30de63efd25
D
de
C
{ "A": 0.06827998593912847, "B": 0.05504920602612491, "C": 0.6871226160748762, "D": 0.18954819195987035 }
2.952197
0c57a6ec-3bd6-4e0e-b537-b530c59f1dff
A
en
A
{ "A": 0.872980261804571, "B": 0.01816222452969636, "C": 0.023209156557006543, "D": 0.08564835710872613 }
2.950067
409804c6-59a5-4e2b-bacc-1fefba951497
D
en
D
{ "A": 0.02336939113212468, "B": 0.015432530902296366, "C": 0.024966831305879178, "D": 0.9362312466596998 }
2.957876
a2d986db-39be-4c77-b9f5-42121894f08d
D
de
A
{ "A": 0.32698750737726856, "B": 0.21982756692559077, "C": 0.22244173433494313, "D": 0.2307431913621976 }
2.950634
57ff5b10-a299-4925-bc68-a081babff9e4
D
en
D
{ "A": 0.005729158721468677, "B": 0.0199277370772223, "C": 0.03971082749461625, "D": 0.9346322767066927 }
2.956289
f6ae12ec-8883-46a9-ae23-501fc5a28cc7
D
en
A
{ "A": 0.7483557964473626, "B": 0.11264409449474032, "C": 0.027548316462996797, "D": 0.1114517925949003 }
2.961218
e12d17fc-d693-485d-9f64-819eeb20d0f7
C
de
D
{ "A": 0.3731760405081874, "B": 0.034627327857835254, "C": 0.04406707859279841, "D": 0.5481295530411789 }
2.951386
d07ca0cd-1d74-4cf1-87fa-8fa8daf95f32
A
de
A
{ "A": 0.7171775579898287, "B": 0.15444384277219758, "C": 0.04373897562868024, "D": 0.08463962360929353 }
2.959463
45d52949-17fe-448d-a664-ba16dee317d8
B
en
B
{ "A": 0.011470234231310147, "B": 0.623134799579476, "C": 0.21487450442991407, "D": 0.1505204617592998 }
2.962871
96ffd9f8-abbc-48d0-a9fa-526829477276
B
en
C
{ "A": 0.09337953545751071, "B": 0.25667462275445485, "C": 0.6313349201609556, "D": 0.018610921627078864 }
2.954243
51770c25-1485-44fd-827b-488bb374f81f
D
de
B
{ "A": 0.10069245172522115, "B": 0.48949695928283127, "C": 0.284303772820484, "D": 0.12550681617146353 }
2.957452
38845176-6cba-4009-90d4-aef53270496b
A
de
A
{ "A": 0.5047924395441115, "B": 0.033368480798023376, "C": 0.09209933232572784, "D": 0.3697397473321373 }
2.95863
79e9fe4a-1fe9-4f71-b635-bfa606c58f46
D
de
A
{ "A": 0.5953657855336928, "B": 0.19412676736689685, "C": 0.16176675357090273, "D": 0.048740693528507635 }
2.475965
8e6cbbbb-cbf5-44a2-94a9-e198b74b3539
B
en
B
{ "A": 0.34533151721267624, "B": 0.5927862357060887, "C": 0.0435608580722801, "D": 0.018321389008954928 }
2.475249
011698c7-494e-451b-b875-e368db021fbd
B
de
B
{ "A": 0.04212163915774237, "B": 0.8253881897464656, "C": 0.08589479300761235, "D": 0.04659537808817977 }
2.473349
3fcdadf0-79c9-4563-bb7a-41cc7f83b393
D
de
D
{ "A": 0.06961261480898115, "B": 0.06623662893374328, "C": 0.21342982283532871, "D": 0.6507209334219469 }
2.473549
ae3de186-5d2a-46cb-a3ac-152e1b7a055a
D
de
D
{ "A": 0.09298039852939957, "B": 0.2011550915070529, "C": 0.10741825838906974, "D": 0.5984462515744778 }
2.477374
93b57328-79cd-4181-8e47-83afcf9d0807
C
de
A
{ "A": 0.5610906444829301, "B": 0.14539648327734306, "C": 0.0961707942088435, "D": 0.1973420780308833 }
2.473959
02b7efac-35c6-4dd6-bf99-dc584e28a12d
C
en
C
{ "A": 0.004770870845901573, "B": 0.4001496893514645, "C": 0.5910868857857787, "D": 0.003992554016855251 }
2.472676
0d857dac-ac07-4c80-911b-cebfdf58c611
C
en
D
{ "A": 0.07846947713450204, "B": 0.04795462225396931, "C": 0.3529209308370564, "D": 0.5206549697744722 }
2.473925
e4304e83-7865-4168-8501-ca4cd97f791a
C
en
C
{ "A": 0.024249776969410095, "B": 0.015603350030661734, "C": 0.9555718258379629, "D": 0.004575047161965264 }
2.47276
bb038739-550e-4a8f-a798-b1415b21525e
D
de
C
{ "A": 0.04779614417759501, "B": 0.1837687297967074, "C": 0.7209837902876429, "D": 0.04745133573805473 }
2.476864
79af4ef6-4a47-4892-af87-f484c1929e00
D
de
B
{ "A": 0.31775432987191177, "B": 0.331028829551621, "C": 0.20637386655275378, "D": 0.1448429740237135 }
2.476633
56ac59ae-e39d-442a-aa9f-1465e2cd04fa
A
de
C
{ "A": 0.06510015373656929, "B": 0.10066786627286768, "C": 0.800699394151639, "D": 0.03353258583892409 }
2.479
1c763207-1b94-4dfa-ac5c-f247b7945f36
D
de
C
{ "A": 0.2378286146449177, "B": 0.170999564591506, "C": 0.4330431344971608, "D": 0.15812868626641544 }
2.483352
08892416-a3f9-4db2-b883-63275d1d5ded
B
de
C
{ "A": 0.02475060073686484, "B": 0.03950569589242557, "C": 0.90820391743417, "D": 0.027539785936539534 }
2.487116
1354e3b2-87d7-4bb6-9f70-7144f8c317a4
C
de
C
{ "A": 0.17567781818527906, "B": 0.1182157849435306, "C": 0.6559484707709938, "D": 0.05015792610019655 }
2.487032
94789d49-b59d-4fd9-9ad4-a96424e3d1de
B
en
A
{ "A": 0.8212125695537907, "B": 0.05522250663838078, "C": 0.03177543301752004, "D": 0.09178949079030854 }
2.485649
20612e7b-f053-4375-a561-d2c7e11ab28f
D
de
C
{ "A": 0.014811310044541776, "B": 0.1868574647898661, "C": 0.555746847230525, "D": 0.2425843779350671 }
2.484874
01ee2974-6644-4b9a-84f7-9fdde5e3b29a
B
de
D
{ "A": 0.1720513864118972, "B": 0.21430057010278428, "C": 0.24488835537556852, "D": 0.36875968810975 }
2.488547
57a77037-e740-4dfa-8ed1-1408a9c43572
D
de
C
{ "A": 0.18445972262110868, "B": 0.14517551897301656, "C": 0.4739113811335374, "D": 0.1964533772723372 }
2.487849
f0c2944f-ee83-4261-b4d7-805ecc81963b
B
en
B
{ "A": 0.0015580700351669253, "B": 0.9890006300886752, "C": 0.0061343645889275255, "D": 0.0033069352872302227 }
2.488423
138e801d-8173-4f14-bed9-7a1be73e4db6
A
de
A
{ "A": 0.9977821734206929, "B": 0.000605246891415722, "C": 0.00122710039641377, "D": 0.00038547929147761075 }
2.484988
ae992f44-2b14-466c-ab7a-d72ed4686b28
A
de
D
{ "A": 0.26678148305433025, "B": 0.11046185571717036, "C": 0.08290705431944224, "D": 0.5398496069090571 }
2.484025
38783200-42e6-465a-98b8-410a27eb4f56
A
en
B
{ "A": 0.004171105226341916, "B": 0.9459434002950303, "C": 0.03750384280779014, "D": 0.012381651670837518 }
2.490743
f56cb1d8-2473-406d-a2bd-0d304a7f018a
C
en
C
{ "A": 0.02020531854324114, "B": 0.042752351688202135, "C": 0.8458353159575799, "D": 0.09120701381097679 }
2.487425
7c35c6fa-a334-4838-b06a-d04039530093
C
en
C
{ "A": 0.006460515405053965, "B": 0.16158977445049374, "C": 0.5466725355291915, "D": 0.2852771746152608 }
2.487212
21c4bf96-9105-4828-868a-68d1fff30bf2
C
en
B
{ "A": 0.0800580182800436, "B": 0.5669829913771293, "C": 0.24266599769881764, "D": 0.11029299264400948 }
2.486559
1e259900-340f-407e-a2da-e2648d2b3ddf
C
de
D
{ "A": 0.3227713812812551, "B": 0.022584006385582635, "C": 0.21840033224591535, "D": 0.4362442800872469 }
2.483874
e61cb406-b329-4fcf-8799-ff319e77c2df
A
de
D
{ "A": 0.21873419553892293, "B": 0.1165498102694787, "C": 0.04712766930653722, "D": 0.6175883248850611 }
2.49087
714317b4-2c6f-44a9-96b5-fef1c6b9583e
C
de
C
{ "A": 0.16494683755311565, "B": 0.040964529001893395, "C": 0.5983629107771954, "D": 0.19572572266779548 }
2.485301
550a5a54-4b73-46c6-86cb-42ced01a9eac
C
de
C
{ "A": 0.30744320754018367, "B": 0.240011071966435, "C": 0.3545730113537433, "D": 0.09797270913963796 }
2.488373
b8eeab4c-42df-44d0-8779-44f24c7e710a
D
de
C
{ "A": 0.237067242235766, "B": 0.13314599091286708, "C": 0.5031437068092249, "D": 0.126643060042142 }
2.488546
75e76320-801e-4bb2-846a-15ef2f8b1eb3
B
de
C
{ "A": 0.16926865468407126, "B": 0.02038824714264369, "C": 0.7594086071379974, "D": 0.05093449103528753 }
2.488158
7620ea1c-6625-408c-a0dd-8ebce1a1bc3e
D
en
B
{ "A": 0.01887273133656022, "B": 0.48219436998216536, "C": 0.1351256220832824, "D": 0.36380727659799195 }
2.456337
855390ce-a070-47f5-9e53-90fe865aebf1
A
de
C
{ "A": 0.04801969443563053, "B": 0.03795809639352099, "C": 0.5430890312121291, "D": 0.37093317795871933 }
2.455422
f0945cdc-4d18-4667-a385-cb4e7c216399
B
de
B
{ "A": 0.07892806316031799, "B": 0.4176086877385061, "C": 0.2786052551007915, "D": 0.2248579940003844 }
2.455836
7326d38e-7515-4e58-963d-63a7160652d4
D
en
A
{ "A": 0.36910162188544193, "B": 0.28365391787520106, "C": 0.053486356934834, "D": 0.29375810330452307 }
2.455433
02736bf2-f16b-4a56-8ca8-7235c47869cc
A
de
A
{ "A": 0.6012287639111795, "B": 0.12964149695407673, "C": 0.1627671367999344, "D": 0.10636260233480926 }
2.456988
6140e65b-e69c-44f1-911b-9756e23c90fd
C
de
B
{ "A": 0.08439715010489633, "B": 0.4297751206781835, "C": 0.2292232245483921, "D": 0.256604504668528 }
2.459083
8849a57f-e14e-4a6e-802f-cd203874abf1
C
en
C
{ "A": 0.03936527860694506, "B": 0.012364834609853867, "C": 0.9397766837469863, "D": 0.008493203036214671 }
2.456197
5b380ec2-39c0-4f82-bb3d-056b45bdf8b1
A
de
A
{ "A": 0.9226540481878324, "B": 0.010815451215445235, "C": 0.026853643639567285, "D": 0.03967685695715501 }
2.452006
4424a2b6-0ade-4618-8cc7-0b0a9c5f2188
B
de
B
{ "A": 0.07455347682441874, "B": 0.775183619409189, "C": 0.11213351266524221, "D": 0.038129391101150094 }
2.455967
70a2ae1a-ae96-431d-920b-82eddc7106a4
D
de
D
{ "A": 0.012311212431530888, "B": 0.01372273330500225, "C": 0.2663929205741523, "D": 0.7075731336893145 }
2.455993
6ed7021e-47b7-42ec-8076-9126f05a282a
B
de
B
{ "A": 0.3169665249668889, "B": 0.5296893276347873, "C": 0.09662287524586065, "D": 0.056721272152463136 }
2.455361
57d99688-a9c5-4733-9ae9-1bda7f986081
C
en
C
{ "A": 0.037479114317141154, "B": 0.07892933217077072, "C": 0.8334099159034202, "D": 0.050181637608668006 }
2.458548
1941019a-ebc2-45ef-842a-8bec34e10d46
A
de
A
{ "A": 0.976639179087905, "B": 0.015640694891646965, "C": 0.004778110113587423, "D": 0.002942015906860505 }
2.457417
4bd0a5ca-0294-4872-af1d-56bcccca74d9
C
de
C
{ "A": 0.1939351884218458, "B": 0.27021338189619964, "C": 0.5206175122643356, "D": 0.015233917417618963 }
2.484843
e916509f-647a-4114-b34c-d021c7201cf7
B
de
B
{ "A": 0.20266651480076045, "B": 0.4346165307442271, "C": 0.20214360087110675, "D": 0.16057335358390573 }
2.483213
20a01998-8cc0-4c7a-b45d-2f36719ebb5a
C
de
D
{ "A": 0.015981270544118735, "B": 0.032593023494053505, "C": 0.04413858176409272, "D": 0.907287124197735 }
2.48743
End of preview.

Decision 3.0 on LEXam (mcq_4_choices)

vllm-sr/d3 scores 61.75 (1,022 / 1,655) on the LEXam-Benchmark/LEXam mcq_4_choices test split.

Protocol One /v1/systemone request per item with a single choice question (lettered options with their text); prediction = argmax of the returned option probabilities. No sampling, no chain of thought.
Items All 1,655; 1,655 answered, 0 errors
Model vllm-sr/d3 @ cbff3a3cb0b074b6585fe9f747b9f9f5557a0e75, BF16
Dataset LEXam-Benchmark/LEXam @ 31e57ae395f92ed7284a4c28d278c55da892b898
Serving vLLM 0.32.0rc1.dev190+g3f1bcfd82 (docker.io/vllm/vllm-openai@sha256:bf73124820b8fd80aaca193721ed02dda278f55ce464cd421d83fbbbb8505f0b), --max-model-len 8192 --enforce-eager --gpu-memory-utilization 0.8
Hardware 1x NVIDIA A100 SXM4 80GB
Date 2026-10-17

By language: German 50.8, English 80.1.

Files

  • results/d3/lexam_mcq4.jsonl: per-item gold, prediction, option probabilities and latency
  • results/d3/lexam_mcq4.summary.json: totals and the sha256 of the results file
  • scripts/run_mc.py: the scorer (--bench lexam_mcq4); scripts/serve.sh: the vLLM launch
  • eval_results/vllm-sr__d3/.eval_results/lexam.yaml: the entry proposed to the model repo

Reproduce 

MODEL_ROOT=<dir containing d3-bf16/> scripts/serve.sh d3-bf16 vllm-sr/d3 \
  --max-model-len 8192 --enforce-eager --gpu-memory-utilization 0.8
python scripts/run_mc.py --bench lexam_mcq4 --model vllm-sr/d3 --out results/d3/lexam_mcq4.jsonl

run_mc.py reads the dataset from scripts/data/LEXam/mcq_4_choices/ (download the pinned revision there first).
Downloads last month
20
Total file size:
403 kB