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
traj_id: string
turn_idx: int64
prefix: list<item: struct<role: string, content: string>>
child 0, item: struct<role: string, content: string>
child 0, role: string
child 1, content: string
reference_turn: string
suffix_len: int64
n_prefix_chars: int64
prev_manifest: null
shards: list<item: struct<active: bool, key: string, n_turns: int64, sha256: string>>
child 0, item: struct<active: bool, key: string, n_turns: int64, sha256: string>
child 0, active: bool
child 1, key: string
child 2, n_turns: int64
child 3, sha256: string
created_at: string
corpus_epoch: int64
to
{'corpus_epoch': Value('int64'), 'created_at': Value('string'), 'prev_manifest': Value('null'), 'shards': List({'active': Value('bool'), 'key': Value('string'), 'n_turns': Value('int64'), 'sha256': 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 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
traj_id: string
turn_idx: int64
prefix: list<item: struct<role: string, content: string>>
child 0, item: struct<role: string, content: string>
child 0, role: string
child 1, content: string
reference_turn: string
suffix_len: int64
n_prefix_chars: int64
prev_manifest: null
shards: list<item: struct<active: bool, key: string, n_turns: int64, sha256: string>>
child 0, item: struct<active: bool, key: string, n_turns: int64, sha256: string>
child 0, active: bool
child 1, key: string
child 2, n_turns: int64
child 3, sha256: string
created_at: string
corpus_epoch: int64
to
{'corpus_epoch': Value('int64'), 'created_at': Value('string'), 'prev_manifest': Value('null'), 'shards': List({'active': Value('bool'), 'key': Value('string'), 'n_turns': Value('int64'), '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.
Affine SN120 turn corpus — epoch 1 (unverified mirror)
Mirror of the public Affine (Bittensor netuid 120) turn corpus shard, fetched from the Hippius S3 bucket on 2026-08-07.
- Source manifest:
https://s3.hippius.com/affine-sn120/turns/manifest.json - Source shard:
https://s3.hippius.com/affine-sn120/turns/shards/turns_epoch_0001.jsonl.gz corpus_epoch: 1 ·n_turns(declared): 9000 · turns (counted): 9000- Per-turn schema:
traj_id,turn_idx,prefix,reference_turn,suffix_len,n_prefix_chars
⚠️ Integrity warning — this data does NOT match the pinned manifest hash
The Affine corpus contract is sha-pinned and fail-closed: the eval pod verifies every shard's sha256 against the manifest and refuses to use a corpus that does not match. This mirror fails that check and must be treated as unverified / untrusted:
| sha256 | |
|---|---|
manifest pins (turns_epoch_0001.jsonl.gz) |
a30a3b5da61186a200dd2e16e4bc3e4bdf1a6ffaeaf81e005b5f8121ba5608c9 |
| this file (downloaded 2026-08-07) | 902d8dd9b24ee9dd439175e9c19bc7a12c03b7d57e012474af4aec317be73b91 |
The download was complete and valid gzip (HTTP 200, application/gzip, full
Content-Length, decompresses to 9000 JSONL lines), and the hash was stable across
repeated fetches — so this is not a truncated or corrupt download. The turn count
matches exactly, suggesting a re-serialization / re-compression of the same underlying
turns rather than different content. However, because the sha256 disagrees with the
official manifest pin, do not treat this as the canonical SN120 corpus. Anyone
relying on the pinned corpus should fetch and verify directly from the source manifest.
Files
turns/shards/turns_epoch_0001.jsonl.gz— the raw shard as downloadedmanifest.json— the source manifest (for reference / hash comparison)
- Downloads last month
- 35