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
example_id: string
context_id: string
task_id: string
pair_index: int64
input: list<item: list<item: int64>>
child 0, item: list<item: int64>
child 0, item: int64
output: list<item: list<item: int64>>
child 0, item: list<item: int64>
child 0, item: int64
model_profile: string
artifact_sha256: string
transform: string
color_map: null
completion_tokens: int64
partition: string
anchor: string
demonstrations: list<item: struct<input: list<item: list<item: int64>>, output: list<item: list<item: int64>>>>
child 0, item: struct<input: list<item: list<item: int64>>, output: list<item: list<item: int64>>>
child 0, input: list<item: list<item: int64>>
child 0, item: list<item: int64>
child 0, item: int64
child 1, output: list<item: list<item: int64>>
child 0, item: list<item: int64>
child 0, item: int64
query_indices: list<item: int64>
child 0, item: int64
artifact: string
profile_valid: bool
demonstration_indices: list<item: int64>
child 0, item: int64
finish_reason: string
heldout_pair_indices: list<item: int64>
child 0, item: int64
source_path: string
mix_name: string
variant: int64
parent_ids: list<item: string>
child 0, item: string
prompt_tokens: int64
to
{'context_id': Value('string'), 'task_id': Value('string'), 'partition': Value('string'), 'parent_ids': List(Value('string')), 'mix_name': Value('string'), 'source_path': Value('string'), 'variant': Value('int64'), 'transform': Value('string'), 'color_map': Value('null'), 'demonstration_indices': List(Value('int64')), 'query_indices': List(Value('int64')), 'demonstrations': List({'input': List(List(Value('int64'))), 'output': List(List(Value('int64')))}), 'heldout_pair_indices': List(Value('int64')), 'anchor': Value('string'), 'model_profile': Value('string'), 'artifact': Value('string'), 'profile_valid': Value('bool'), 'prompt_tokens': Value('int64'), 'completion_tokens': Value('int64'), 'finish_reason': Value('string'), 'artifact_sha256': 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
example_id: string
context_id: string
task_id: string
pair_index: int64
input: list<item: list<item: int64>>
child 0, item: list<item: int64>
child 0, item: int64
output: list<item: list<item: int64>>
child 0, item: list<item: int64>
child 0, item: int64
model_profile: string
artifact_sha256: string
transform: string
color_map: null
completion_tokens: int64
partition: string
anchor: string
demonstrations: list<item: struct<input: list<item: list<item: int64>>, output: list<item: list<item: int64>>>>
child 0, item: struct<input: list<item: list<item: int64>>, output: list<item: list<item: int64>>>
child 0, input: list<item: list<item: int64>>
child 0, item: list<item: int64>
child 0, item: int64
child 1, output: list<item: list<item: int64>>
child 0, item: list<item: int64>
child 0, item: int64
query_indices: list<item: int64>
child 0, item: int64
artifact: string
profile_valid: bool
demonstration_indices: list<item: int64>
child 0, item: int64
finish_reason: string
heldout_pair_indices: list<item: int64>
child 0, item: int64
source_path: string
mix_name: string
variant: int64
parent_ids: list<item: string>
child 0, item: string
prompt_tokens: int64
to
{'context_id': Value('string'), 'task_id': Value('string'), 'partition': Value('string'), 'parent_ids': List(Value('string')), 'mix_name': Value('string'), 'source_path': Value('string'), 'variant': Value('int64'), 'transform': Value('string'), 'color_map': Value('null'), 'demonstration_indices': List(Value('int64')), 'query_indices': List(Value('int64')), 'demonstrations': List({'input': List(List(Value('int64'))), 'output': List(List(Value('int64')))}), 'heldout_pair_indices': List(Value('int64')), 'anchor': Value('string'), 'model_profile': Value('string'), 'artifact': Value('string'), 'profile_valid': Value('bool'), 'prompt_tokens': Value('int64'), 'completion_tokens': Value('int64'), 'finish_reason': Value('string'), 'artifact_sha256': 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.
RuleLoopViT NVARC TaskSketch 20K
Incrementally generated grounded text contexts and grid-transition examples for RuleLoopViT GL training.
Intended size
- 20,000 parent-disjoint
nvarc_trainingtasks. - 10 text contexts per task (approximately 200,000 contexts).
- Up to 10 query pairs per context (approximately 2,000,000 examples).
Files
data/contexts/shard-*.jsonl.gz: generated TaskSketch artifacts, exact measured evidence, demonstrations, transformations, and provenance.data/examples/shard-*.jsonl.gz: query input/output pairs keyed bycontext_id.data/shards/shard-*.json: checksums and validation statistics.progress.json: current uploaded-shard progress.
Each shard is uploaded atomically and is immutable after successful upload. Generation can resume from the first absent shard on another machine.
Evaluation protection
- The historical 100-task
nvarc_unseen_devpanel is parent-disjoint from all 20,000 tasks. - A stratified 200-task seen-task panel reserves two original pairs per task.
- Reserved pairs are excluded from training targets, grounding demonstrations, text-generation demonstrations, and augmentation sources.
Context variations
Every selected task receives the eight D4 geometric views plus two deterministic global color permutations. Qwen3.5-9B receives 2–3 transformed demonstrations, rendered images, compact grid serialization, and deterministic measured evidence. Thinking is disabled. The generated seven-field concept vector is stored together with the measured evidence consumed by GL.
- Downloads last month
- 1,715