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
frozen_feature_recipe: struct<development_used: bool, final_normalization: string, hidden_dim: int64, hidden_layer: string, (... 234 chars omitted)
child 0, development_used: bool
child 1, final_normalization: string
child 2, hidden_dim: int64
child 3, hidden_layer: string
child 4, hidden_token: string
child 5, learned_normalization: bool
child 6, projected_features: string
child 7, projection_dim: int64
child 8, projection_rng: string
child 9, projection_seed: int64
child 10, representation_model: string
child 11, scalar_features: list<item: string>
child 0, item: string
child 12, version: string
gae_lambda: double
gamma: double
kl: double
max_context: int64
new_external_calls: int64
nu: double
old_probs_posthoc_replayed: bool
policy_optimizer_steps: int64
records: list<item: struct<actual_sampling_logprobs_archived: bool, candidate_id: string, cost_critic_values: (... 622 chars omitted)
child 0, item: struct<actual_sampling_logprobs_archived: bool, candidate_id: string, cost_critic_values: list<item: (... 610 chars omitted)
child 0, actual_sampling_logprobs_archived: bool
child 1, candidate_id: string
child 2, cost_critic_values: list<item: double>
child 0, item: double
child 3, current_logprobs: list<item: double>
child 0, item: double
child 4, old_logprobs: list<item: double>
child 0, item: double
child 5, raw_cm: double
child 6, raw_rm: double
child 7, reference_last_hidden: list<item: double>
child 0, item: double
child 8, reference_logprobs: list<item: double>
child 0, item: double
child 9, response_length: int64
child 10, response_token_ids: list<item: int64>
child 0, item: int64
child 11, reward_critic_values: list<item: double>
child 0, item: double
child 12, tensor_binding: struct<attention_mask: struct<dtype: string, sha256: string, shape: list<item: int64>>, input_ids: s (... 142 chars omitted)
child 0, attention_mask: struct<dtype: string, sha256: string, shape: list<item: int64>>
child 0, dtype: string
child 1, sha256: string
child 2, shape: list<item: int64>
child 0, item: int64
child 1, input_ids: struct<dtype: string, sha256: string, shape: list<item: int64>>
child 0, dtype: string
child 1, sha256: string
child 2, shape: list<item: int64>
child 0, item: int64
child 2, pixel_values: struct<dtype: string, sha256: string, shape: list<item: int64>>
child 0, dtype: string
child 1, sha256: string
child 2, shape: list<item: int64>
child 0, item: int64
child 13, tensor_file: string
child 14, tensor_file_sha256: string
local_png_pixels_and_hashes_checked: int64
image_hash_recipe: string
no_upload_or_api_requests_executed: bool
to
{'records': List({'kind': Value('string'), 'record_index': Value('int64'), 'record_id': Value('string'), 'source_id': Value('string'), 'question_sha256': Value('string'), 'response_sha256': Value('string'), 'legacy_image_sha256': Value('string'), 'canonical_image_sha256': Value('string'), 'original_png_sha256': Value('string'), 'public_train_candidates': List({'row_index': Value('int64'), 'source_id': Value('string')}), 'public_sample_value_comparison': Value('string'), 'source_file': Value('string'), 'request_id': Value('string'), 'source_client_metadata': {'policy_round': Value('int64'), 'candidate_id': Value('int64'), 'source_train_row': Value('int64'), 'image_png_sha256': Value('string'), 'hide_from_judge': Value('string')}}), 'image_hash_recipe': Value('string'), 'local_png_pixels_and_hashes_checked': Value('int64'), 'no_upload_or_api_requests_executed': Value('bool')}
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 544, 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 3052, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2559, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2584, 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 558, in _iter_arrow
for key, pa_table in iterator:
^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 440, 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 361, 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 2381, in table_cast
return cast_table_to_schema(table, schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2309, in cast_table_to_schema
raise CastError(
...<3 lines>...
)
datasets.table.CastError: Couldn't cast
frozen_feature_recipe: struct<development_used: bool, final_normalization: string, hidden_dim: int64, hidden_layer: string, (... 234 chars omitted)
child 0, development_used: bool
child 1, final_normalization: string
child 2, hidden_dim: int64
child 3, hidden_layer: string
child 4, hidden_token: string
child 5, learned_normalization: bool
child 6, projected_features: string
child 7, projection_dim: int64
child 8, projection_rng: string
child 9, projection_seed: int64
child 10, representation_model: string
child 11, scalar_features: list<item: string>
child 0, item: string
child 12, version: string
gae_lambda: double
gamma: double
kl: double
max_context: int64
new_external_calls: int64
nu: double
old_probs_posthoc_replayed: bool
policy_optimizer_steps: int64
records: list<item: struct<actual_sampling_logprobs_archived: bool, candidate_id: string, cost_critic_values: (... 622 chars omitted)
child 0, item: struct<actual_sampling_logprobs_archived: bool, candidate_id: string, cost_critic_values: list<item: (... 610 chars omitted)
child 0, actual_sampling_logprobs_archived: bool
child 1, candidate_id: string
child 2, cost_critic_values: list<item: double>
child 0, item: double
child 3, current_logprobs: list<item: double>
child 0, item: double
child 4, old_logprobs: list<item: double>
child 0, item: double
child 5, raw_cm: double
child 6, raw_rm: double
child 7, reference_last_hidden: list<item: double>
child 0, item: double
child 8, reference_logprobs: list<item: double>
child 0, item: double
child 9, response_length: int64
child 10, response_token_ids: list<item: int64>
child 0, item: int64
child 11, reward_critic_values: list<item: double>
child 0, item: double
child 12, tensor_binding: struct<attention_mask: struct<dtype: string, sha256: string, shape: list<item: int64>>, input_ids: s (... 142 chars omitted)
child 0, attention_mask: struct<dtype: string, sha256: string, shape: list<item: int64>>
child 0, dtype: string
child 1, sha256: string
child 2, shape: list<item: int64>
child 0, item: int64
child 1, input_ids: struct<dtype: string, sha256: string, shape: list<item: int64>>
child 0, dtype: string
child 1, sha256: string
child 2, shape: list<item: int64>
child 0, item: int64
child 2, pixel_values: struct<dtype: string, sha256: string, shape: list<item: int64>>
child 0, dtype: string
child 1, sha256: string
child 2, shape: list<item: int64>
child 0, item: int64
child 13, tensor_file: string
child 14, tensor_file_sha256: string
local_png_pixels_and_hashes_checked: int64
image_hash_recipe: string
no_upload_or_api_requests_executed: bool
to
{'records': List({'kind': Value('string'), 'record_index': Value('int64'), 'record_id': Value('string'), 'source_id': Value('string'), 'question_sha256': Value('string'), 'response_sha256': Value('string'), 'legacy_image_sha256': Value('string'), 'canonical_image_sha256': Value('string'), 'original_png_sha256': Value('string'), 'public_train_candidates': List({'row_index': Value('int64'), 'source_id': Value('string')}), 'public_sample_value_comparison': Value('string'), 'source_file': Value('string'), 'request_id': Value('string'), 'source_client_metadata': {'policy_round': Value('int64'), 'candidate_id': Value('int64'), 'source_train_row': Value('int64'), 'image_png_sha256': Value('string'), 'hide_from_judge': Value('string')}}), 'image_hash_recipe': Value('string'), 'local_png_pixels_and_hashes_checked': Value('int64'), 'no_upload_or_api_requests_executed': Value('bool')}
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.
Joint experiment pool, responses, budget and request indices
Actual first delta payload: 3690463 bytes in four data files. Includes32 numeric pool records (logprobs/critic values/reference features and retained tensor shape/dtype/hash bindings),64 owned generated responses,120 original completed ledger attempts, and160 N8 request/calibration index records. These overlap with earlier experiments and are not independent human labels. No API requests or training were run.
The SQLite ledger was opened read-only/immutable after confirming no journal/WAL, and its exact two-table schema was checked. Its120 COMPLETE attempts do not mean the1250 cap was consumed or prove independent billing. The original database hash and unchanged-source check are recorded.
The32 request client-only records restore candidate_id/source_train_row/policy_round/pnghash/protocol-context indices omitted by the previous cautious historical-feedback projection. Their protocol label client_only is not an archival prohibition. The128 calibration indices originate from a historical training-derived1024/256 split, not official590 validation. Public-train index candidates are preserved, but actual pixel/text comparisons for those160 are still explicitly pending; references alone are not claimed as reconstruction proof. Related official64 policy-cache bindings were independently verified elsewhere.
Recover
hf download wangzyuan/SafeRLHFV-Joint-Feedback-Pool-Ledger-Rescue --repo-type dataset --revision COMMIT --local-dir ./joint-delta
Pin the commit from its publication receipt, verify output hashes in RECOVERY_MANIFEST, and read JSON. Full referenced tensor_file payloads, original questions/images and other remaining experiment candidates are not all included. Tensor hash/shape/dtype descriptors are preserved rather than mistakenly discarded as raw input/image tensors. Native checkpoint backup completion is separate from this data release.
- Downloads last month
- 128