jhu-clsp/mmBERT-base
Fill-Mask β’ Updated β’ 383k β’ β’ 257
Dataset Viewer
The dataset viewer is not available for this subset.
Cannot get the split names for the config 'default' of the dataset.
Exception: SplitsNotFoundError
Message: The split names could not be parsed from the dataset config.
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.14/site-packages/datasets/inspect.py", line 286, in get_dataset_config_info
for split_generator in builder._split_generators(
~~~~~~~~~~~~~~~~~~~~~~~~~^
StreamingDownloadManager(base_path=builder.base_path, download_config=download_config)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
)
^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 101, in _split_generators
pa_table = next(iter(self._generate_tables(**splits[0].gen_kwargs, allow_full_read=False)))[1]
~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
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 136, in _cast_table
pa_table = table_cast(pa_table, 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 2312, in cast_table_to_schema
cast_array_to_feature(
~~~~~~~~~~~~~~~~~~~~~^
table[name] if name in table_column_names else pa.array([None] * len(table), type=schema.field(name).type),
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
feature,
^^^^^^^^
)
^
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 1861, in wrapper
return pa.chunked_array([func(chunk, *args, **kwargs) for chunk in array.chunks])
~~~~^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2118, in cast_array_to_feature
casted_array_values = _c(array.values, feature.feature)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 1863, in wrapper
return func(array, *args, **kwargs)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2071, in cast_array_to_feature
return pa.StructArray.from_arrays(arrays, names=list(feature), mask=array.is_null())
~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "pyarrow/array.pxi", line 4305, in pyarrow.lib.StructArray.from_arrays
File "pyarrow/array.pxi", line 1852, in pyarrow.lib.Array.validate
check_status(self.ap.Validate())
File "pyarrow/error.pxi", line 92, in pyarrow.lib.check_status
pyarrow.lib.ArrowInvalid: Struct child array #2 invalid: Invalid: Length spanned by list offsets (2) larger than values array (length 1)
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/split_names.py", line 68, in compute_split_names_from_streaming_response
for split in get_dataset_split_names(
~~~~~~~~~~~~~~~~~~~~~~~^
path=dataset,
^^^^^^^^^^^^^
config_name=config,
^^^^^^^^^^^^^^^^^^^
token=hf_token,
^^^^^^^^^^^^^^^
)
^
File "/usr/local/lib/python3.14/site-packages/datasets/inspect.py", line 340, in get_dataset_split_names
info = get_dataset_config_info(
path,
...<6 lines>...
**config_kwargs,
)
File "/usr/local/lib/python3.14/site-packages/datasets/inspect.py", line 291, in get_dataset_config_info
raise SplitsNotFoundError("The split names could not be parsed from the dataset config.") from err
datasets.inspect.SplitsNotFoundError: The split names could not be parsed from the dataset config.Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
mmBERT Pre-training Data P2
π¨ DEPRECATED: KNOWN BUG π¨
This is an old version of the data with a critical bug: token IDs were stored as
uint16and overflow. mmBERT's vocabulary is larger than 65,535, so any token ID above that limit wraps around and is silently corrupted.β Do NOT use this data for training. This card is kept for reference only (data mixture, sources, and documentation).
β Fixed data: orionweller/mmBERT-pretraining-data-chunk0
βΉοΈ Note: there is no fixed version split by source. If you need the per-source breakdown, use the information on this card, but get the actual tokens from the fixed dataset above.
Phase 1 of 3: Diverse multilingual pre-training data mixture (trained for 2.3T tokens) used to train the mmBERT model suite.
NOTE: this is only P2 of the pre-training data due to HF limits, you need to download and combine all three into one folder
This dataset contains the pre-training phase data used to train all mmBERT encoder models. The data is provided in MDS format ready for use with Composer and the ModernBERT training repository.
π Data Composition
| Data Source | Tokens (B) | Percentage | Description |
|---|---|---|---|
| FineWeb2 | 1,196.6 | 60.2% | High-quality multilingual web crawl data |
| DCLM | 600.0 | 30.2% | High-quality English web crawl data |
| Starcoder | 100.6 | 5.1% | Code repositories and files |
| Arxiv | 27.8 | 1.4% | Academic preprints |
| StackExchange | 18.6 | 0.9% | Q&A forums |
| Tulu Flan | 15.3 | 0.8% | Instruction-following data |
| Dolmino Math | 11.2 | 0.6% | Mathematical content |
| PeS2o | 8.4 | 0.4% | Scientific papers |
| Wikipedia (MegaWika) | 4.7 | 0.2% | Encyclopedia articles |
| Books | 4.3 | 0.2% | Literature and reference books |
| StackExchange (Dolmino) | 1.4 | 0.1% | Curated Q&A content |
| Total | 1,989.0 | 100.0% | Diverse mixture for foundation training |
π Language Coverage
This phase covers 60 languages plus code, with an inverse temperature sampling schedule starting at Ο=0.7. Languages include:
- High-resource: English (34.5%), Russian (5.8%), German (4.4%), Spanish (4.5%), French (4.0%), Chinese (5.2%)
- Mid-resource: Italian, Portuguese, Japanese, Dutch, Polish, and 45 others
- Scripts: Latin, Cyrillic, Arabic, Chinese, Japanese, Thai, and many more
π Usage
For pre-training, see the ModernBERT repo: https://github.com/AnswerDotAI/ModernBERT
Direct Access
Use the script at this link to load any section of the dataset on the fly. This will fail if you try to access too many samples though, due to HF rate-limiting. To download the full dataset, use HF Hub's Snapshot Download.
π Related Resources
- Models: mmBERT Model Suite
- Phase 2: Mid-training Data (600B tokens)
- Phase 3: Decay Phase Data (100B tokens)
- Checkpoints: Training Checkpoints
- Paper: Arxiv link
- Code: GitHub Repository
Citation
@misc{marone2025mmbertmodernmultilingualencoder,
title={mmBERT: A Modern Multilingual Encoder with Annealed Language Learning},
author={Marc Marone and Orion Weller and William Fleshman and Eugene Yang and Dawn Lawrie and Benjamin Van Durme},
year={2025},
eprint={2509.06888},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2509.06888},
}
- Downloads last month
- 46,005
