Datasets:
The dataset viewer is not available for this subset.
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/webdataset/webdataset.py", line 80, in _split_generators
raise ValueError(
...<2 lines>...
)
ValueError: The TAR archives of the dataset should be in WebDataset format, but the files in the archive don't share the same prefix or the same types.
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.
Self-conditioning deep reasoning datasets
This repository hosts the original synthetic image pairs and locally derived metadata used by the main experiments in Self-Conditioning Enables Deep Sequential Reasoning in Flow Matching. The datasets preserve their original image payloads, metadata bytes, and sample ordering.
| Directory | Samples | Format |
|---|---|---|
line_tracing/ |
1,000,000 train and 1,000 validation | Packed image shards with metadata; 50 permutations |
line_tracing/nfe_eval/ |
The first 200 original validation samples | Shared validation shard and subset metadata |
long_multiplication/ |
200,000 train and 100 test | Image pairs and metadata |
tinystories/300k_no_edit_384x384/ |
300,000 train and 21,990 test | TinyStories rendered at 384 by 384 pixels |
diffthinker_local/ |
12 formatted CSVs, 40,000 Sudoku-v2 training pairs, and 30,000/300 Jigsaw train/test pairs | Derived metadata and generated image pairs |
Each .tar file is an independent archive. Extract every part for a split into
its parent directory. Do not concatenate archive files.
mkdir -p line_tracing
for f in line_tracing/train.part*.tar line_tracing/val.part*.tar; do
tar -xf "$f" -C line_tracing
done
mkdir -p line_tracing/nfe_eval
for f in line_tracing/nfe_eval/val.part*.tar; do
tar -xf "$f" -C line_tracing/nfe_eval
done
Line-tracing metadata uses references such as shard:000000:0. Its
shards/shard_index.json locates the packed image data. The reproduction
repository reads this format directly and verifies the metadata hashes and row
counts before use. Training uses all one million training samples; the main
evaluation selects the first 20 validation samples. The NFE sweep selects the
first 200 validation samples.
MANIFEST.json lists every archive's size and SHA-256. The reproduction scripts
download a pinned revision, extract the selected archives, and verify the
dataset's integrity. The matching main-experiment checkpoints are available at
alfalfa-w/self-conditioning-deep-reasoning.
The VSP, Maze, TSP, base Sudoku, and edit-image data remain referenced from
internlm/EndoCoT-Data
at revision 34c84daf1403582804c0c1f778813a11e11ec6c9. Evaluation ground truth
comes from InternLM/EndoCoT at commit
8cf17c06a029aa50c91c0867cd3c547d2a91c658. Rendered TinyStories uses
roneneldan/TinyStories text. Jigsaw follows the DiffThinker COCO-image recipe.
The source data retains its applicable rights; this repository does not assert
one blanket license for all data sources.
- Downloads last month
- -