The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
Error code: DatasetGenerationError
Exception: TypeError
Message: int() argument must be a string, a bytes-like object or a real number, not 'NoneType'
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1531, in _prepare_split_single
for key, record 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/webdataset/webdataset.py", line 127, in _generate_examples
for example_idx, example in enumerate(self._get_pipeline_from_tar(tar_path, tar_iterator)):
~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/webdataset/webdataset.py", line 32, in _get_pipeline_from_tar
for filename, f in tar_iterator:
^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/utils/track.py", line 49, in __iter__
for x in self.generator(*self.args):
~~~~~~~~~~~~~~^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/utils/file_utils.py", line 1400, in _iter_from_urlpath
with xopen(urlpath, "rb", download_config=download_config, block_size=0) as f:
~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/utils/file_utils.py", line 977, in xopen
file_obj = fs.open(paths[0], mode)
File "<string>", line 3, in open
File "/usr/local/lib/python3.14/unittest/mock.py", line 1176, in __call__
return self._mock_call(*args, **kwargs)
~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/unittest/mock.py", line 1180, in _mock_call
return self._execute_mock_call(*args, **kwargs)
~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/unittest/mock.py", line 1247, in _execute_mock_call
result = effect(*args, **kwargs)
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 786, in wrapped
tracker.files[urlpath] = {"read": 0, "size": int(f.size)}
~~~^^^^^^^^
TypeError: int() argument must be a string, a bytes-like object or a real number, not 'NoneType'
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 1393, 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 1571, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the datasetNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
rs unknown | __key__ string | __url__ string |
|---|---|---|
[
117,
115,
101,
32,
115,
117,
112,
101,
114,
58,
58,
42,
59,
13,
10,
13,
10,
112,
117,
98,
40,
99,
114,
97,
116,
101,
41,
32,
102,
110,
32,
98,
105,
116,
40,
99,
58,
32,
85,
50,
53,
54,
44,
32,
105,
58,
32,
117,
115,
10... | arith/adder | hf://datasets/yudduy/yukonbench@4247fc98ea78771163e8fa6d9dbfa8a419be7514/solutions/ecdsa-fail/00dc611283bdceac5fbad7a605100329925282cc28b83c2038ea5dd4aa929732.tar.gz |
[
117,
115,
101,
32,
115,
117,
112,
101,
114,
58,
58,
42,
59,
13,
10,
13,
10,
112,
117,
98,
40,
99,
114,
97,
116,
101,
41,
32,
102,
110,
32,
99,
109,
112,
95,
108,
116,
95,
105,
110,
116,
111,
95,
102,
97,
115,
116,
40,
98... | arith/compare | hf://datasets/yudduy/yukonbench@4247fc98ea78771163e8fa6d9dbfa8a419be7514/solutions/ecdsa-fail/00dc611283bdceac5fbad7a605100329925282cc28b83c2038ea5dd4aa929732.tar.gz |
"dXNlIHN1cGVyOjoqOw0KDQojW2lubGluZV0NCmZuIG1hajFfaW5wdXRzX2Rpc3RpbmN0KGE6IFF1Yml0SWQsIGs6IFF1Yml0SWQ(...TRUNCATED) | arith/const_arith | "hf://datasets/yudduy/yukonbench@4247fc98ea78771163e8fa6d9dbfa8a419be7514/solutions/ecdsa-fail/00dc6(...TRUNCATED) |
"DQp1c2Ugc3VwZXI6Oio7DQoNCm1vZCBhZGRlcjsNCm1vZCBjb21wYXJlOw0KbW9kIGNvbnN0X2FyaXRoOw0KbW9kIG1vZHVsYXI(...TRUNCATED) | arith/mod | "hf://datasets/yudduy/yukonbench@4247fc98ea78771163e8fa6d9dbfa8a419be7514/solutions/ecdsa-fail/00dc6(...TRUNCATED) |
"DQp1c2Ugc3VwZXI6Oio7DQoNCnB1YihjcmF0ZSkgZm4gbW9kX2FkZF9xcShiOiAmbXV0IEIsIGFjYzogJltRdWJpdElkXSwgYTo(...TRUNCATED) | arith/modular | "hf://datasets/yudduy/yukonbench@4247fc98ea78771163e8fa6d9dbfa8a419be7514/solutions/ecdsa-fail/00dc6(...TRUNCATED) |
"DQp1c2Ugc3VwZXI6Oio7DQoNCiNbYWxsb3coZGVhZF9jb2RlKV0NCnB1YihjcmF0ZSkgZm4gbW9kX211bF93cml0ZV9pbnRvX3p(...TRUNCATED) | arith/multiply | "hf://datasets/yudduy/yukonbench@4247fc98ea78771163e8fa6d9dbfa8a419be7514/solutions/ecdsa-fail/00dc6(...TRUNCATED) |
"dXNlIHN1cGVyOjoqOw0KDQpwdWIoY3JhdGUpIGZuIGFkZF9uYml0X3FxX2Zhc3QoYjogJm11dCBCLCBhOiAmW1F1Yml0SWRdLCB(...TRUNCATED) | arith/nbit | "hf://datasets/yudduy/yukonbench@4247fc98ea78771163e8fa6d9dbfa8a419be7514/solutions/ecdsa-fail/00dc6(...TRUNCATED) |
"Ly8gQVVUTy1HRU5FUkFURUQgLSBkZWVwLXN0cmlwIGRlYWQtQ0NYIHNldCBmb3IgdGhlIEdBUCBkZWx0YT0yIGNpcmN1aXQuDQo(...TRUNCATED) | d2_deep_strip | "hf://datasets/yudduy/yukonbench@4247fc98ea78771163e8fa6d9dbfa8a419be7514/solutions/ecdsa-fail/00dc6(...TRUNCATED) |
"Ly8gQVVUTy1HRU5FUkFURUQgYnkgY2Vuc3VzX21pbmVyOiBpZGVudGl0eS1rZXllZCBkZWVwLXN0cmlwIHRhYmxlLg0KLy8gc2h(...TRUNCATED) | deep_strip_keys | "hf://datasets/yudduy/yukonbench@4247fc98ea78771163e8fa6d9dbfa8a419be7514/solutions/ecdsa-fail/00dc6(...TRUNCATED) |
"Ly8hIENsYXNzaWNhbCBjb252ZXJnZW5jZSBwcmUtZmlsdGVyIGZvciBkaWFsb2ctR0NEIEZpYXQtU2hhbWlyIGlzbGFuZCBzZWF(...TRUNCATED) | dialog_gcd_classical_filter | "hf://datasets/yudduy/yukonbench@4247fc98ea78771163e8fa6d9dbfa8a419be7514/solutions/ecdsa-fail/00dc6(...TRUNCATED) |
YukonBench research records
Experiment records from YukonBench, which replays historical research challenges. The code repository contains the runner, verifier rules, and task definitions. This dataset contains the inputs, results, and source needed to check individual runs.
Finding a run
Browse by challenge, episode, model, and harness:
runs/ecdsa-fail/e-late/
configs/<snapshot-hash>.json
baselines/keep-start/run-01/
gpt-6.1-sol/codex/run-01/
glm-5.3/pi/run-01/
glm-5.3/oh-my-pi/run-01/
deepseek-v4-pro/pi/run-01/
summary.json
events.jsonl
transcript.jsonl
solutions/ecdsa-fail/<content-hash>.tar.gz
Each model/harness folder contains repeated trials. Run numbers identify published trials;
seeds, start times, and original campaign identities remain in their summaries. Multi-agent
strategies have a multi-agent/ subfolder beneath the harness.
configs/stores each exact input snapshot once: settings, pinned episode, and frozen prices. Each summary links to its snapshot throughconfig_file.summary.jsonrecords the outcome, measurements, software versions, platform, and flags.events.jsonlrecords charges, budget clocks, injections, evaluations, keyed checks, and acceptance decisions.transcript.jsonlcontains agent messages, tool activity, and research notes, with agent and turn identities retained.solutions/contains compressed source submitted or captured during a run, stored once per content hash. These are agent outputs; the challenge and verifier live in GitHub. The hash covers sorted relative paths and SHA-256 of each file's bytes (see code).
Reading the results
E-late is a development comparison, not a held-out benchmark. Costs use frozen token and CPU prices in normalized USD, including subscription usage. Keyed integrity checks are diagnostic and do not decide official acceptance.
The main branch contains the original three baseline campaigns and completed, eligible model experiments. Keep-start retains the starting source; provided-solution adopts the public injection; historical-solutions replays known reference source. Their budgets differ from the model trials. Model trials share the episode and budget schedule, but launch batches used different price sheets and wall caps. Read each run's input snapshot and software flags before comparing results. The model grid remains in progress. Zero improvement is still a result.
Infrastructure checks, smoke tests, probes, exploratory runs, failed infrastructure attempts,
exclusions, and replaced attempts are preserved at
history-2026-10-04
(revision eaf03ae305d4f1c09902a2fe1b51312c7e5c6ee5). Use that revision to audit attrition
and retry costs; main alone does not contain all attempted runs. Original input hashes,
Git revisions, identities, and measurements remain unchanged.
Personal and runtime paths are made portable. Credentials, copied harness settings, and session databases are excluded; missing diagnostics are listed explicitly.
Downloading records
Pin a revision to retrieve the same files later:
hf download yudduy/yukonbench --repo-type dataset --revision <revision> --local-dir records
yb report records/runs/ecdsa-fail/e-late
yb report produces Markdown and CSV locally from the underlying records. The dataset
contains no duplicate generated reports.
Solution source comes from the pinned ECDSA.fail challenge and participating agents. See the pinned upstream repository for licensing and attribution.
- Downloads last month
- 238