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
README.md: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
RECOVERY_MANIFEST.json: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
TERMS.md: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
records/calibration192-generated-responses.jsonl: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
records/core-endtoend16-v1/base64-feedback-v1/LEDGER_SNAPSHOT.json: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
records/core-endtoend16-v1/base64-recovery-diagnostic-send-v3/LEDGER_SNAPSHOT.json: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
records/core-endtoend16-v1/base64-recovery-diagnostic-send-v3b/LEDGER_SNAPSHOT.json: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
records/core-endtoend16-v1/base64-recovery-send-v2/LEDGER_SNAPSHOT.json: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
records/execution-20261020-v1/signal-220/verified-labels-221/LABELS.json: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
records/fresh-feedback-v2/sequential-v1/LEDGER_SNAPSHOT.json: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
records/full32-measurement-feedback-v1/FULL32_LABELS.json: struct<bytes: int64, sha256: s
...
int64
schema_version: int64
source_file_count: int64
feedback_is_human_gold: bool
restore: string
independent_blind_evaluation: bool
source_files: list<item: struct<source_relative_path: string, source_bytes: int64, source_sha256: string, source_u (... 300 chars omitted)
child 0, item: struct<source_relative_path: string, source_bytes: int64, source_sha256: string, source_unchanged_st (... 288 chars omitted)
child 0, source_relative_path: string
child 1, source_bytes: int64
child 2, source_sha256: string
child 3, source_unchanged_stat: bool
child 4, output: string
child 5, removed_fields: struct<receipt: int64, source_ledger: int64, source_file: int64, question: int64, image_file: int64, (... 104 chars omitted)
child 0, receipt: int64
child 1, source_ledger: int64
child 2, source_file: int64
child 3, question: int64
child 4, image_file: int64
child 5, tensor_file: int64
child 6, sensitive_pattern:response: int64
child 7, response_token_ids_for_withheld_response: int64
child 6, records: int64
child 7, output_row_index: int64
outputs: list<item: struct<path: string, bytes: int64, sha256: string>>
child 0, item: struct<path: string, bytes: int64, sha256: string>
child 0, path: string
child 1, bytes: int64
child 2, sha256: string
purpose: string
payload_files: int64
raw_images_original_questions_original_api_receipts_included: bool
to
{'schema_version': Value('int64'), 'purpose': Value('string'), 'source_files': List({'source_relative_path': Value('string'), 'source_bytes': Value('int64'), 'source_sha256': Value('string'), 'source_unchanged_stat': Value('bool'), 'output': Value('string'), 'removed_fields': {'receipt': Value('int64'), 'source_ledger': Value('int64'), 'source_file': Value('int64'), 'question': Value('int64'), 'image_file': Value('int64'), 'tensor_file': Value('int64'), 'sensitive_pattern:response': Value('int64'), 'response_token_ids_for_withheld_response': Value('int64')}, 'records': Value('int64'), 'output_row_index': Value('int64')}), 'outputs': List({'path': Value('string'), 'bytes': Value('int64'), 'sha256': Value('string')}), 'source_file_count': Value('int64'), 'payload_files': Value('int64'), 'payload_bytes': Value('int64'), 'generated_records': Value('int64'), 'feedback_is_human_gold': Value('bool'), 'independent_blind_evaluation': Value('bool'), 'raw_images_original_questions_original_api_receipts_included': Value('bool'), 'projection': {'excluded_keys': List(Value('string')), 'secret_fields_excluded': List(Value('string')), 'sensitive_strings': Value('string'), 'source_files_unchanged': Value('bool')}, 'coverage_limits': List(Value('string')), 'restore': Value('string')}
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
README.md: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
RECOVERY_MANIFEST.json: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
TERMS.md: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
records/calibration192-generated-responses.jsonl: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
records/core-endtoend16-v1/base64-feedback-v1/LEDGER_SNAPSHOT.json: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
records/core-endtoend16-v1/base64-recovery-diagnostic-send-v3/LEDGER_SNAPSHOT.json: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
records/core-endtoend16-v1/base64-recovery-diagnostic-send-v3b/LEDGER_SNAPSHOT.json: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
records/core-endtoend16-v1/base64-recovery-send-v2/LEDGER_SNAPSHOT.json: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
records/execution-20261020-v1/signal-220/verified-labels-221/LABELS.json: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
records/fresh-feedback-v2/sequential-v1/LEDGER_SNAPSHOT.json: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
records/full32-measurement-feedback-v1/FULL32_LABELS.json: struct<bytes: int64, sha256: s
...
int64
schema_version: int64
source_file_count: int64
feedback_is_human_gold: bool
restore: string
independent_blind_evaluation: bool
source_files: list<item: struct<source_relative_path: string, source_bytes: int64, source_sha256: string, source_u (... 300 chars omitted)
child 0, item: struct<source_relative_path: string, source_bytes: int64, source_sha256: string, source_unchanged_st (... 288 chars omitted)
child 0, source_relative_path: string
child 1, source_bytes: int64
child 2, source_sha256: string
child 3, source_unchanged_stat: bool
child 4, output: string
child 5, removed_fields: struct<receipt: int64, source_ledger: int64, source_file: int64, question: int64, image_file: int64, (... 104 chars omitted)
child 0, receipt: int64
child 1, source_ledger: int64
child 2, source_file: int64
child 3, question: int64
child 4, image_file: int64
child 5, tensor_file: int64
child 6, sensitive_pattern:response: int64
child 7, response_token_ids_for_withheld_response: int64
child 6, records: int64
child 7, output_row_index: int64
outputs: list<item: struct<path: string, bytes: int64, sha256: string>>
child 0, item: struct<path: string, bytes: int64, sha256: string>
child 0, path: string
child 1, bytes: int64
child 2, sha256: string
purpose: string
payload_files: int64
raw_images_original_questions_original_api_receipts_included: bool
to
{'schema_version': Value('int64'), 'purpose': Value('string'), 'source_files': List({'source_relative_path': Value('string'), 'source_bytes': Value('int64'), 'source_sha256': Value('string'), 'source_unchanged_stat': Value('bool'), 'output': Value('string'), 'removed_fields': {'receipt': Value('int64'), 'source_ledger': Value('int64'), 'source_file': Value('int64'), 'question': Value('int64'), 'image_file': Value('int64'), 'tensor_file': Value('int64'), 'sensitive_pattern:response': Value('int64'), 'response_token_ids_for_withheld_response': Value('int64')}, 'records': Value('int64'), 'output_row_index': Value('int64')}), 'outputs': List({'path': Value('string'), 'bytes': Value('int64'), 'sha256': Value('string')}), 'source_file_count': Value('int64'), 'payload_files': Value('int64'), 'payload_bytes': Value('int64'), 'generated_records': Value('int64'), 'feedback_is_human_gold': Value('bool'), 'independent_blind_evaluation': Value('bool'), 'raw_images_original_questions_original_api_receipts_included': Value('bool'), 'projection': {'excluded_keys': List(Value('string')), 'secret_fields_excluded': List(Value('string')), 'sensitive_strings': Value('string'), 'source_files_unchanged': Value('bool')}, 'coverage_limits': List(Value('string')), 'restore': 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.
SafeRLHF-V feedback and calibration rescue
Actual owned research data, not a code-only package: 202 source files project into 11 data files / 7,114,422 bytes. This includes 192 generated Qwen response records with their response token IDs, historical log-probability arrays and 2560-dimensional reference hidden vectors; 192 calibration feedback/features and input reconstruction identifiers; 32 older measurement feedback labels; and six budget snapshots. The 192-record tables overlap and must not be added as independent populations.
Original questions and images, private filesystem references, API response envelopes and service receipt bodies are excluded. Withheld references have hashes, and each source has explicit omitted-field counts and its original hash in RECOVERY_MANIFEST.json. Source hashes identify originals, NOT a claim that all original source bytes are remotely preserved. Exactly one of the 192 response strings matched a sensitive-content pattern and is withheld together with its response token IDs; 191 response strings remain. Its metadata, original response hash and explicit omission counts are retained. Outputs are reviewed projections, not byte-identical copies of whole source JSONs.
Feedback is model/proxy feedback, not human gold. Calibration/development data are not independent blind evaluation. Preserve original actual_sampling_logprobs_archived, old_logprobs_posthoc_replayed, optimizer-step and feedback-call fields; no new training or quality acceptance is claimed. Budget reservations/snapshots are not proof of charged network calls. There are no model/optimizer state tensors in this dataset.
Recovery
hf download wangzyuan/SafeRLHFV-Feedback-Calibration-Rescue --repo-type dataset --revision COMMIT --local-dir ./feedback-rescue
Use the fixed commit recorded by the publication receipt. Verify the 11 output hashes/byte sizes listed in RECOVERY_MANIFEST.json, read JSON/JSONL and join index, request_id, source_id and the source/image/question hashes. Questions/images remain a separate authorized data dependency. Numeric features are not raw-image backups. Reproduction requires the matching source/transforms/model and original datasets; this package alone cannot certify exact training continuation.
- Downloads last month
- 59