The dataset viewer is not available for this split.
Error code: StreamingRowsError
Exception: TypeError
Message: Couldn't cast array of type string to null
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 2315, 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 2121, 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
_c(children[name] if name in array_fields else null_array, subfeature)
~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
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 2155, in cast_array_to_feature
return array_cast(
array,
...<2 lines>...
allow_decimal_to_str=allow_decimal_to_str,
)
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 2014, in array_cast
raise TypeError(f"Couldn't cast array of type {_short_str(array.type)} to {_short_str(pa_type)}")
TypeError: Couldn't cast array of type string to nullNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
TGCL: extracted concepts and test states
The concept banks and test-set hidden states behind the tables of TGCL (token-grounded concept learning for vision-language models), so the tables can be rebuilt without extracting concepts again. Code and instructions: github.com/Mak-Ta-Reque/TGCL-LVLM.
Model google/gemma-3n-E4B-it (hidden states after the final norm,
model.language_model.norm, 2048 dimensions); datasets COCO-10 (the 10 most
frequent COCO objects) and ImageNet-1k (20 training / 10 test images per
class); methods TGCL (SAM segments, sliding windows), CoX-LMM, CRAFT, MCD;
decompositions semi-NMF, SAE, SIMPLE; seeds 1–3.
Use
git clone -b release https://github.com/Mak-Ta-Reque/TGCL-LVLM && cd TGCL-LVLM
huggingface-cli download abka03/TGCL-LVLM --repo-type dataset --local-dir artifacts
tar -xzf artifacts/evaluation/coco10.tar.gz -C artifacts/evaluation
python scripts/evaluate_banks.py table --out artifacts/evaluation/coco10
The repository README describes every command (concept-quality table, re-evaluation on the GPU, extraction from the beginning).
Contents
concepts/<dataset>/<method>_seed<s>/concept/<decomposition>/combined_concept_*.pth
concept bank (torch.load): concepts [K, 2048], concept_names, text_grounding
(words), image_grounding_paths / _bboxes / _masks (training segments)
test_states/coco10/seed<s>/, test_states/imagenet1k/
explanations[_<setting>]/snmf/vlm_explanations.json answers, per-token concepts
explanations[_<setting>]/snmf/vlm_explanations_activations.npy hidden states [N, 2048]
explanations[_<setting>]/snmf/vlm_explanations_center.npy calibration mean
settings: MCQ (no suffix), image_only, text_only, combined
evaluation/<dataset>.tar.gz unpacks to evaluation/<dataset>/<run>/:
bank_eval.json, test_accuracy.json, eval[_<setting>]/<d>/concept_curves.json,
clip_bert_topk_*.csv, word_alignment_table.csv (per-run results of the tables)
tables/
concept_quality.csv, coco10_test_results.csv, imagenet1k_test_results.csv
TGCL's banks are stored as learned; the evaluation applies TGCL's combination (symmetric alignment by Löwdin orthogonalization, noise removal keeping 50 % of each concept's coordinates).
License
MIT (see LICENSE), © 2026 Md Abdul Kadir. No images are included: image
paths point into the COCO and ImageNet datasets (data/..., built by the
repository's dataset scripts), whose images remain under their own terms of
use. The hidden states come from Gemma-3n, whose use is governed by the Gemma
Terms of Use.
- Downloads last month
- 115