The dataset viewer is not available for this split.
Error code: StreamingRowsError
Exception: TypeError
Message: Couldn't cast array of type int64 to null
Traceback: Traceback (most recent call last):
File "/src/services/worker/src/worker/utils.py", line 149, in get_rows_or_raise
return get_rows(
dataset=dataset,
...<4 lines>...
column_names=column_names,
)
File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
return func(*args, **kwargs)
File "/src/services/worker/src/worker/utils.py", line 129, in get_rows
rows_plus_one = list(itertools.islice(safe_iter(ds, dataset=dataset), rows_max_number + 1))
File "/src/services/worker/src/worker/utils.py", line 489, in safe_iter
yield from ds.decode(False) if ds.features else ds
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2818, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2355, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2380, in _iter_arrow
for key, pa_table in self.ex_iterable._iter_arrow():
~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 536, in _iter_arrow
for key, pa_table in iterator:
^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 419, in _iter_arrow
for key, pa_table in self.generate_tables_fn(**gen_kwags):
~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
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 2369, in table_cast
return cast_table_to_schema(table, schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2303, in cast_table_to_schema
cast_array_to_feature(
~~~~~~~~~~~~~~~~~~~~~^
table[name] if name in table_column_names else pa.array([None] * len(table), type=schema.field(name).type),
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
feature,
^^^^^^^^
)
^
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 1852, in wrapper
return pa.chunked_array([func(chunk, *args, **kwargs) for chunk in array.chunks])
~~~~^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2059, in cast_array_to_feature
_c(array.field(name) if name in array_fields else null_array, subfeature)
~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 1854, in wrapper
return func(array, *args, **kwargs)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2143, in cast_array_to_feature
return array_cast(
array,
...<2 lines>...
allow_decimal_to_str=allow_decimal_to_str,
)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 1854, in wrapper
return func(array, *args, **kwargs)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2005, in array_cast
raise TypeError(f"Couldn't cast array of type {_short_str(array.type)} to {_short_str(pa_type)}")
TypeError: Couldn't cast array of type int64 to nullNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
CooperBench flash_10 evaluation traces
Every run here is the same frozen contract: subset flash_10 (10 pairs / 10 tasks / 9 repos),
agent mini_swe_agent_v2, setting coop + --git, Modal sandboxes, step_limit=200,
max_model_len=49152. Produced by scripts/eval_e2e.py in
CooperTrain. Two runs are comparable only if their
run_meta.json agree — check it rather than assuming.
Each run directory contains, per pair:
| file | contents |
|---|---|
agentN_traj.json |
one agent's trajectory, post-compaction window only |
agentN_full_traj.json |
the full history under segments[]; use kind=="solver" (the summarizer segments are near-duplicates and double-count) |
agentN.patch |
the graded artifact — the diff taken from that agent's PR |
eval.json |
merge status and per-feature pass/fail |
result.json |
agent exit statuses, duration, message counts |
summary.json |
pairs and features passed |
run_meta.json |
the full contract this run used |
Submission is a PR, not a patch file. Since cooperbench 0.0.28 the agent opens its own PR and
grading reads git ls-remote --tags team refs/tags/pr/<agent>. The log line
no patch.txt to publish is a legacy message, not a failure — an empty patch means no PR was
opened. Before reading a 0/10 as a model result, grep the trajectories for gh: command not found.
The runs
Model baselines — no injected context
| run | model | pairs | features |
|---|---|---|---|
flash10-qwen3-5-9b, BASELINE-9b-noctx-backup |
Qwen3.5-9B | 2/10 | 7/20 |
flash10-qwen3-6-35b-a3b |
Qwen3.6-35B-A3B | 2/10 | 7/20 |
flash10-qwen3-5-122b |
Qwen3.5-122B-A10B | 3/10 | 10/20 |
The 27B base (4/10) was overwritten: eval_e2e.py derives the run name from the model, so the
27B ceiling run reused flash10-qwen3-5-27b. That directory now holds the 9/10 ceiling. The script
now refuses to overwrite a finished run.
Teacher-context ceilings
Each injects a block into every agent's own feature.md, then restores the originals. The
question each answers is what a teacher could know that a student could not.
| run | injected context | pairs | features |
|---|---|---|---|
flash10-qwen3-5-27b |
protocol + role + the benchmark's gold patch | 9/10 | 18/20 |
ceiling-ctx-9b |
same, on the 9B | 5/10 | 11/20 |
ceiling-protocol-only-27b |
per-task workflows — no diff block, but still names files, line numbers and hidden-test content |
7/10 | 16/20 |
ceiling-protocol-only-9b |
same, on the 9B | 1/10 | 8/20 |
ceiling-backlog-27b |
shared-backlog protocol — generic, no task content | 4/10 | 13/20 |
ceiling-negotiate-27b |
negotiate-your-own — a menu of 10 protocols plus a decision rule, no task content | 4/10 | 13/20 |
The headline: every context that beats base contains the answer. Both purely procedural
protocols score exactly what the unbriefed model scores — while demonstrably working
(ceiling-negotiate-27b produced 7 identical merges out of 10, the most of any run, and
ceiling-backlog-27b produced 10 clean merges with 0 conflicts).
On-policy distillation students
LoRA r=32 over Qwen3.5-9B, trained by coopertrain/opd/. Base to beat is 2/10, 7/20.
| run | teacher | pairs | features |
|---|---|---|---|
flash10-qwen3-5-9b-opd-r1 |
27B + gold patch, 1 epoch | 0/10 | 3/20 |
flash10-qwen3-5-9b-opd-self |
same 9B weights + gold patch, 1 epoch | 0/10 | 1/20 |
flash10-qwen3-5-9b-opd-proto |
27B + per-task workflows, 3 epochs | 1/10 | 4/20 |
Void — do not use
VOID-opd-self-gh-broken ran against cooperbench 0.0.22 rather than 0.0.28: eval_e2e.py
invoked the CLI by name and PATH resolved it to a second install. 0.0.22 names the shared daemon
team, leaves real GitHub as origin, installs no gh shim and grades cat patch.txt. All 19
agents hit gh: command not found, pushed correct commits to team/agentN, and were scored empty.
Kept only as a worked example of the failure.
- Downloads last month
- 12