The dataset viewer is not available for this split.
Error code: StreamingRowsError
Exception: CastError
Message: Couldn't cast
family: string
gate: string
items: list<item: struct<name: string, prompt: string, intermediates: list<item: string>, source: string, c (... 122 chars omitted)
child 0, item: struct<name: string, prompt: string, intermediates: list<item: string>, source: string, consistency: (... 110 chars omitted)
child 0, name: string
child 1, prompt: string
child 2, intermediates: list<item: string>
child 0, item: string
child 3, source: string
child 4, consistency: double
child 5, gate_variant: string
child 6, eval_render: string
child 7, readout: struct<kind: string, offsets: list<item: int64>>
child 0, kind: string
child 1, offsets: list<item: int64>
child 0, item: int64
eval_render_note: string
to
{'family': Value('string'), 'gate': Value('string'), 'items': List({'name': Value('string'), 'prompt': Value('string'), 'intermediates': List(Value('string')), 'consistency': Value('float64'), 'consistency_strict': Value('float64'), 'source': Value('string'), 'gate_variant': Value('string'), 'eval_render': Value('string'), 'readout': {'kind': Value('string'), 'offsets': List(Value('int64'))}, 'probe_token_lens': {'intermediates': List(Value('int64'))}})}
because column names don't match
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 2297, in cast_table_to_schema
raise CastError(
...<3 lines>...
)
datasets.table.CastError: Couldn't cast
family: string
gate: string
items: list<item: struct<name: string, prompt: string, intermediates: list<item: string>, source: string, c (... 122 chars omitted)
child 0, item: struct<name: string, prompt: string, intermediates: list<item: string>, source: string, consistency: (... 110 chars omitted)
child 0, name: string
child 1, prompt: string
child 2, intermediates: list<item: string>
child 0, item: string
child 3, source: string
child 4, consistency: double
child 5, gate_variant: string
child 6, eval_render: string
child 7, readout: struct<kind: string, offsets: list<item: int64>>
child 0, kind: string
child 1, offsets: list<item: int64>
child 0, item: int64
eval_render_note: string
to
{'family': Value('string'), 'gate': Value('string'), 'items': List({'name': Value('string'), 'prompt': Value('string'), 'intermediates': List(Value('string')), 'consistency': Value('float64'), 'consistency_strict': Value('float64'), 'source': Value('string'), 'gate_variant': Value('string'), 'eval_render': Value('string'), 'readout': {'kind': Value('string'), 'offsets': List(Value('int64'))}, 'probe_token_lens': {'intermediates': List(Value('int64'))}})}
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.
Oracle-Lens Behavioral Eval Banks — final
Frozen, pre-registered eval item banks for the oracle-lens project (MATS, Neel Nanda
stream). 17 banks / 1,645 items: 10 base families (multihop, multilingual,
order-ops, poetry, typo, association, directed-modulation, basic-readout,
user-modeling, sandbagging) plus -mt multi-token variants for 7 of them.
Each file is lens-eval-<family>.json with an items list. Items carry an
eval_render mode — most families are rendered as plain text with NO chat
template; directed-modulation (chat_dm), basic-readout implicit prediction
(chat), user-modeling (chat_context), and sandbagging (chat_sandbag) are the
chat exceptions. Always render via
global_workspace.oracle_lens_evals.render_for_eval from the source repo so this
cannot be gotten wrong.
Source of truth: evals/oracle_lens/final/ in the global-workspace repo (which also
holds the staging intermediates, generation sources, and the gen → check → finalize
pipeline scripts under scripts/oracle_lens_evals/). This upload mirrors that
directory verbatim.
⚠️ These are eval questions. Please do not include them in model training corpora.
- Downloads last month
- 28