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
derivation: struct<algorithm: string, packing_counters: string, parent: string, parent_manifest_sha256: string, (... 133 chars omitted)
child 0, algorithm: string
child 1, packing_counters: string
child 2, parent: string
child 3, parent_manifest_sha256: string
child 4, parent_rows: int64
child 5, reason: string
child 6, rows: int64
child 7, rows_copied_verbatim: bool
child 8, selection: string
child 9, sidecar: string
child 10, valid_tokens: int64
files: list<item: struct<bytes: int64, path: string, rows: int64, sha256: string>>
child 0, item: struct<bytes: int64, path: string, rows: int64, sha256: string>
child 0, bytes: int64
child 1, path: string
child 2, rows: int64
child 3, sha256: string
format: string
packing: struct<alignment_padding_tokens: int64, appended_boundary_tokens: int64, full_sequences: int64, over (... 440 chars omitted)
child 0, alignment_padding_tokens: int64
child 1, appended_boundary_tokens: int64
child 2, full_sequences: int64
child 3, overlength_documents_excluded: int64
child 4, overlength_tokens_excluded: int64
child 5, packed_documents: int64
child 6, packed_token_slots: int64
child 7, record_tail_padding_tokens: int64
child 8, sampled_documents: int64
child 9, sampled_out_documents: int64
child 10, sampled_text_utf8_bytes: int64
child 11, smoke_sequences: int64
child 12, source_eos_documents_excluded: int64
child 13, source_eos_tokens_excluded: int64
child 14, source_ro
...
0, item: struct<bytes: int64, path: string, sha256: string>
child 0, bytes: int64
child 1, path: string
child 2, sha256: string
child 3, revision: string
child 4, root: string
child 5, split: string
child 6, text_column: string
tokenizer: struct<files: list<item: struct<bytes: int64, name: string, sha256: string>>, model_id: string, revi (... 340 chars omitted)
child 0, files: list<item: struct<bytes: int64, name: string, sha256: string>>
child 0, item: struct<bytes: int64, name: string, sha256: string>
child 0, bytes: int64
child 1, name: string
child 2, sha256: string
child 1, model_id: string
child 2, revision: string
child 3, runtime: struct<add_bos_token: bool, add_eos_token: bool, added_vocab_entries: int64, bos_token_id: int64, ch (... 216 chars omitted)
child 0, add_bos_token: bool
child 1, add_eos_token: bool
child 2, added_vocab_entries: int64
child 3, bos_token_id: int64
child 4, chat_template_sha256: string
child 5, eos_token_id: int64
child 6, is_fast: bool
child 7, length: int64
child 8, mask_token_id: int64
child 9, max_token_id: int64
child 10, model_max_length: int64
child 11, pad_token_id: int64
child 12, tokenizer_class: string
child 13, vocab_size_attr: int64
training_files: struct<full: list<item: string>>
child 0, full: list<item: string>
child 0, item: string
manifest_sha256: string
to
{'format': Value('string'), 'manifest_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
derivation: struct<algorithm: string, packing_counters: string, parent: string, parent_manifest_sha256: string, (... 133 chars omitted)
child 0, algorithm: string
child 1, packing_counters: string
child 2, parent: string
child 3, parent_manifest_sha256: string
child 4, parent_rows: int64
child 5, reason: string
child 6, rows: int64
child 7, rows_copied_verbatim: bool
child 8, selection: string
child 9, sidecar: string
child 10, valid_tokens: int64
files: list<item: struct<bytes: int64, path: string, rows: int64, sha256: string>>
child 0, item: struct<bytes: int64, path: string, rows: int64, sha256: string>
child 0, bytes: int64
child 1, path: string
child 2, rows: int64
child 3, sha256: string
format: string
packing: struct<alignment_padding_tokens: int64, appended_boundary_tokens: int64, full_sequences: int64, over (... 440 chars omitted)
child 0, alignment_padding_tokens: int64
child 1, appended_boundary_tokens: int64
child 2, full_sequences: int64
child 3, overlength_documents_excluded: int64
child 4, overlength_tokens_excluded: int64
child 5, packed_documents: int64
child 6, packed_token_slots: int64
child 7, record_tail_padding_tokens: int64
child 8, sampled_documents: int64
child 9, sampled_out_documents: int64
child 10, sampled_text_utf8_bytes: int64
child 11, smoke_sequences: int64
child 12, source_eos_documents_excluded: int64
child 13, source_eos_tokens_excluded: int64
child 14, source_ro
...
0, item: struct<bytes: int64, path: string, sha256: string>
child 0, bytes: int64
child 1, path: string
child 2, sha256: string
child 3, revision: string
child 4, root: string
child 5, split: string
child 6, text_column: string
tokenizer: struct<files: list<item: struct<bytes: int64, name: string, sha256: string>>, model_id: string, revi (... 340 chars omitted)
child 0, files: list<item: struct<bytes: int64, name: string, sha256: string>>
child 0, item: struct<bytes: int64, name: string, sha256: string>
child 0, bytes: int64
child 1, name: string
child 2, sha256: string
child 1, model_id: string
child 2, revision: string
child 3, runtime: struct<add_bos_token: bool, add_eos_token: bool, added_vocab_entries: int64, bos_token_id: int64, ch (... 216 chars omitted)
child 0, add_bos_token: bool
child 1, add_eos_token: bool
child 2, added_vocab_entries: int64
child 3, bos_token_id: int64
child 4, chat_template_sha256: string
child 5, eos_token_id: int64
child 6, is_fast: bool
child 7, length: int64
child 8, mask_token_id: int64
child 9, max_token_id: int64
child 10, model_max_length: int64
child 11, pad_token_id: int64
child 12, tokenizer_class: string
child 13, vocab_size_attr: int64
training_files: struct<full: list<item: string>>
child 0, full: list<item: string>
child 0, item: string
manifest_sha256: string
to
{'format': Value('string'), 'manifest_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.
DART FineCode deterministic 1/8 sample — inclusionAI/LLaDA2.0-mini
This is a byte-preserving publication of the sealed training dataset used by
the DART partial-order experiments. It contains 1,713,961 packed rows. The tokenizer is
inclusionAI/LLaDA2.0-mini, sequence length is 2,048, and document starts are
aligned to the training block size of 64.
MANIFEST.json is the authoritative provenance and file inventory. Its
SHA-256 is 45e250f9f4a5302ef1bb04c16577eb8b6d85996368fe1dc359b86465cdb03176. COMPLETE.json binds that exact
manifest. The source release contains 18 files and
46,653,342,155 bytes before publication metadata.
Compatible dependency sidecars
attention(pred-debias5 corrected attention):zimplex/dllm-attention-parents-llada2-finecode-pred-debias5-b64-v1(manifestf4dbab97f495e23682f26ac085b19a90e9e6a15cf664b789d9efb0f23566cd3e)counterfactual(corrected d1-marginal, window 128, tau 2):zimplex/dllm-effect-parents-llada2-finecode-d1marginal-w128-tau2-b64-v3(manifest391fa0f852451457ba2812a8633af7799f18193e5e0d2a489432ccc7c3d67c43)
PAIRINGS.json provides the same mapping in machine-readable form. A sidecar
is keyed by the exact packed input_ids + segment_ids; do not retokenize,
repack, reorder, or rewrite the JSONL payloads.
This is a local-file training release, not a conventional streaming Hugging
Face dataset. Download it to shared storage and verify SHA256SUMS before use.
Token IDs are reversible model inputs and are not an anonymization mechanism.
Absolute paths inside the manifest are historical provenance only.
- Downloads last month
- 28