The dataset viewer is not available for this split.
Error code: StreamingRowsError
Exception: ValueError
Message: Expected object or value
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 291, in _generate_tables
pa_table = paj.read_json(
io.BytesIO(batch), read_options=paj.ReadOptions(block_size=block_size)
)
File "pyarrow/_json.pyx", line 364, in pyarrow._json.read_json
File "pyarrow/error.pxi", line 155, in pyarrow.lib.pyarrow_internal_check_status
File "pyarrow/error.pxi", line 92, in pyarrow.lib.check_status
raise convert_status(status)
pyarrow.lib.ArrowInvalid: JSON parse error: Column() changed from object to string in row 0
During handling of the above exception, another exception occurred:
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 306, in _generate_tables
batch = json_encode_fields_in_json_lines(original_batch, json_field_paths)
File "/usr/local/lib/python3.14/site-packages/datasets/utils/json.py", line 111, in json_encode_fields_in_json_lines
examples = [ujson_loads(line) for line in original_batch.splitlines()]
~~~~~~~~~~~^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/utils/json.py", line 20, in ujson_loads
return pd.io.json.ujson_loads(*args, **kwargs)
~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
ValueError: Expected object or valueNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
Modeldiffr bench
Does Modeldiffr find real changes, stay quiet on fake ones, and catch edited reports? Every audit below uses the lite suite and the released modeldiffr CLI.
- Planted weight changes detected: 4/4 (95% CI 51% to 100%)
- False alarms on identical and near identical copies: 0/4 (95% CI 0% to 49%)
- Real fine tunes, merges and edits flagged as changed: 4/4
- Tampered reports rejected by
modeldiffr verify: 24/24, untouched reports accepted: 12/12
Controls on Qwen2.5 0.5B Instruct
Identical copies count as a false alarm if any metric changes. Near identical copies (noise at 0.0001 of each tensor's std) really are different weights, so their output KL shifts by a tiny but real amount. For them a false alarm means a capability or safety metric moved.
| audit | planted change | expected | result | metrics that changed |
|---|---|---|---|---|
| null-identity-s0 | null identity | no_change | clean | none |
| null-identity-s1 | null identity | no_change | clean | none |
| null-micro-noise-s0 | null micro noise | no_change | clean | output_divergence_kl |
| null-micro-noise-s1 | null micro noise | no_change | clean | output_divergence_kl |
| planted-noise-0.10 | planted noise | change | detected | xstest_unsafe_compliance_rate, or_bench_toxic_compliance_rate, output_divergence_kl |
| planted-noise-0.40 | planted noise | change | detected | hellaswag_ece, gsm8k_accuracy, mgsm_accuracy, ifeval_prompt_accuracy, xstest_safe_over_refusal_rate, or_bench_hard_over_refusal_rate, xstest_unsafe_compliance_rate, or_bench_toxic_compliance_rate, output_divergence_kl |
| planted-prune-0.30 | planted prune | change | detected | gsm8k_accuracy, or_bench_hard_over_refusal_rate, output_divergence_kl |
| planted-zero-last-mlp | planted zero | change | detected | output_divergence_kl |
Real derivatives
| audit | kind | base | derivative | result | largest significant changes |
|---|---|---|---|---|---|
| real-shortthink-sft | sft | Qwen/Qwen3-1.7B | AmareshHebbar/ShortThink_Qwen3_1.7B_SFT | detected | ifeval_prompt_accuracy +0.167 [+0.033, +0.300]; arc_challenge_ece +0.038 [+0.012, +0.079]; output_divergence_kl +0.016 [+0.015, +0.017] |
| real-toolsure-orpo | orpo | Qwen/Qwen2.5-1.5B-Instruct | AmareshHebbar/ToolSure_Qwen2.5_1.5B_ORPO | detected | output_divergence_kl +0.008 [+0.006, +0.009] |
| real-instruct-qwen2.5-0.5b | instruction tuning | Qwen/Qwen2.5-0.5B | Qwen/Qwen2.5-0.5B-Instruct | detected | xstest_unsafe_compliance_rate -0.800 [-0.933, -0.633]; or_bench_toxic_compliance_rate -0.800 [-0.933, -0.667]; xstest_safe_over_refusal_rate +0.400 [+0.233, +0.567] |
| real-bitnarrow-qwen7b | weight edit | Qwen/Qwen2.5-7B-Instruct | qwen7b_narrow | detected | xstest_unsafe_compliance_rate +0.200 [+0.067, +0.367]; output_divergence_kl +0.002 [+0.002, +0.003] |
Tamper test
Each finished report was copied and edited three ways: one metric value moved by 0.01, the verdict flipped, and one number in an item file changed. Then modeldiffr verify ran on the copy.
| edit | tried | rejected |
|---|---|---|
| metric | 12 | 12 |
| verdict | 12 | 12 |
Reproduce
git clone https://github.com/amareshhebbar/modeldiffr && cd modeldiffr
pip install -e ".[hf,tasks,batch]" -r bench/requirements.txt
bash bench/run.sh
Verify without a GPU
python -c "from huggingface_hub import snapshot_download as s; s('AmareshHebbar/Modeldiffr_Bench', repo_type='dataset', local_dir='bench/results')"
python bench/verify.py
Manifest root: ea20b6aaff8a76be6598f5ff92616da638d30083ac2f9817d011e0b7b6c7cf25
- Downloads last month
- 61