Dataset Viewer
The dataset viewer is not available for this split.
Cannot load the dataset split (in streaming mode) to extract the first rows.
Error code: StreamingRowsError
Exception: CastError
Message: Couldn't cast
config: string
job: string
trial: string
reward: double
tests: string
agent_execution_sec: double
status: string
failure_bucket: string
p1_anticheat_read: struct<status: string, evidence: string>
child 0, status: string
child 1, evidence: string
p2_anticheat_write: struct<status: string, evidence: string>
child 0, status: string
child 1, evidence: string
p4_agentic_floor: struct<status: string, evidence: string>
child 0, status: string
child 1, evidence: string
p5_format_vs_reasoning: struct<status: string, evidence: string>
child 0, status: string
child 1, evidence: string
p6_memorization: struct<status: string, evidence: string>
child 0, status: string
child 1, evidence: string
p7_tool_breakdown: struct<execute: int64, edit: int64, read: int64, fetch: int64, think: int64, other: int64>
child 0, execute: int64
child 1, edit: int64
child 2, read: int64
child 3, fetch: int64
child 4, think: int64
child 5, other: int64
p8_struggle_vs_wrong: struct<repeat_exec: int64, rewrites: struct<exterior.py: int64, solve.py: int64, sectors.py: int64, (... 50 chars omitted)
child 0, repeat_exec: int64
child 1, rewrites: struct<exterior.py: int64, solve.py: int64, sectors.py: int64, run.py: int64>
child 0, exterior.py: int64
child 1, solve.py: int64
child 2, sectors.py: int64
child 3, run.py: int64
child 2, reversals: int64
child 3, verdict: string
p10_verbatim_final_statement: string
p14_filesystem_pollution: struct<status: string, evidence: string>
child 0, status: string
child 1, evidence: string
cross_run_concurrence: string
egress_note: string
to
{'config': Value('string'), 'job': Value('string'), 'trial': Value('string'), 'reward': Value('float64'), 'tests': Value('string'), 'agent_execution_sec': Value('float64'), 'status': Value('string'), 'failure_bucket': Value('string'), 'p1_anticheat_read': {'status': Value('string'), 'evidence': Value('string')}, 'p2_anticheat_write': {'status': Value('string'), 'evidence': Value('string')}, 'p4_agentic_floor': {'status': Value('string'), 'evidence': Value('string')}, 'p5_format_vs_reasoning': {'status': Value('string'), 'evidence': Value('string')}, 'p6_memorization': {'status': Value('string'), 'evidence': Value('string')}, 'p7_tool_breakdown': {'execute': Value('int64'), 'edit': Value('int64'), 'read': Value('int64'), 'fetch': Value('int64'), 'think': Value('int64'), 'other': Value('int64')}, 'p8_struggle_vs_wrong': {'repeat_exec': Value('int64'), 'rewrites': {'core.py': Value('int64'), 'hierarchy.py': Value('int64'), 'driver.py': Value('int64'), 'pipeline.py': Value('int64'), 'figs.py': Value('int64')}, 'reversals': Value('int64'), 'verdict': Value('string')}, 'p10_verbatim_final_statement': Value('string'), 'p14_filesystem_pollution': {'status': Value('string'), 'evidence': Value('string')}, 'egress_note': Value('string')}
because column names don't match
Traceback: Traceback (most recent call last):
File "/src/services/worker/src/worker/utils.py", line 147, 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 127, 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 483, 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 2840, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2373, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2398, 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 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
config: string
job: string
trial: string
reward: double
tests: string
agent_execution_sec: double
status: string
failure_bucket: string
p1_anticheat_read: struct<status: string, evidence: string>
child 0, status: string
child 1, evidence: string
p2_anticheat_write: struct<status: string, evidence: string>
child 0, status: string
child 1, evidence: string
p4_agentic_floor: struct<status: string, evidence: string>
child 0, status: string
child 1, evidence: string
p5_format_vs_reasoning: struct<status: string, evidence: string>
child 0, status: string
child 1, evidence: string
p6_memorization: struct<status: string, evidence: string>
child 0, status: string
child 1, evidence: string
p7_tool_breakdown: struct<execute: int64, edit: int64, read: int64, fetch: int64, think: int64, other: int64>
child 0, execute: int64
child 1, edit: int64
child 2, read: int64
child 3, fetch: int64
child 4, think: int64
child 5, other: int64
p8_struggle_vs_wrong: struct<repeat_exec: int64, rewrites: struct<exterior.py: int64, solve.py: int64, sectors.py: int64, (... 50 chars omitted)
child 0, repeat_exec: int64
child 1, rewrites: struct<exterior.py: int64, solve.py: int64, sectors.py: int64, run.py: int64>
child 0, exterior.py: int64
child 1, solve.py: int64
child 2, sectors.py: int64
child 3, run.py: int64
child 2, reversals: int64
child 3, verdict: string
p10_verbatim_final_statement: string
p14_filesystem_pollution: struct<status: string, evidence: string>
child 0, status: string
child 1, evidence: string
cross_run_concurrence: string
egress_note: string
to
{'config': Value('string'), 'job': Value('string'), 'trial': Value('string'), 'reward': Value('float64'), 'tests': Value('string'), 'agent_execution_sec': Value('float64'), 'status': Value('string'), 'failure_bucket': Value('string'), 'p1_anticheat_read': {'status': Value('string'), 'evidence': Value('string')}, 'p2_anticheat_write': {'status': Value('string'), 'evidence': Value('string')}, 'p4_agentic_floor': {'status': Value('string'), 'evidence': Value('string')}, 'p5_format_vs_reasoning': {'status': Value('string'), 'evidence': Value('string')}, 'p6_memorization': {'status': Value('string'), 'evidence': Value('string')}, 'p7_tool_breakdown': {'execute': Value('int64'), 'edit': Value('int64'), 'read': Value('int64'), 'fetch': Value('int64'), 'think': Value('int64'), 'other': Value('int64')}, 'p8_struggle_vs_wrong': {'repeat_exec': Value('int64'), 'rewrites': {'core.py': Value('int64'), 'hierarchy.py': Value('int64'), 'driver.py': Value('int64'), 'pipeline.py': Value('int64'), 'figs.py': Value('int64')}, 'reversals': Value('int64'), 'verdict': Value('string')}, 'p10_verbatim_final_statement': Value('string'), 'p14_filesystem_pollution': {'status': Value('string'), 'evidence': Value('string')}, 'egress_note': Value('string')}
because column names don't matchNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
FrontierPhysics PR #36 review trajectories — nonlinear-boson-star-sideband-threshold
Benchmark + rubric-review bundle for benchflow-ai/FrontierPhysics PR #36 at head 6ce04ff4.
pr36-trajectories.zip: review report, static policy results, per-run C0 trajectory audits, 4 rubric review_report.jsons, and per-run evidence (full acp_trajectory.jsonl, result/config/timing, verifier output incl. each run's paper.pdf). Oracle run ships result/timing/verifier stdout only — its solution artifacts are the answer key and are deliberately excluded.
Runs: oracle (reward 1.0, 10/10) + 4x claude-agent-acp / claude-opus-5 (reasoning effort max, no-skill, Daytona) — all 9/10, failing only T10; rubric raw quality 100/100/93.4/100%, all blockers pass. See pr36/report.txt inside the zip for the full review.
- Downloads last month
- 55