Dataset Preview
Duplicate
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed because of a cast error
Error code:   DatasetGenerationCastError
Exception:    DatasetGenerationCastError
Message:      An error occurred while generating the dataset

All the data files must have the same columns, but at some point there are 2 new columns ({'outlier_type', 'alpha'}) and 1 missing columns ({'iter'}).

This happened while the csv dataset builder was generating data using

hf://datasets/Srishti280992/repro-mamba-icl-outliers-actual/actual_outputs/fig2.csv (at revision d0daa81356fd508db339a949d418e462bb5ade6c), ['hf://datasets/Srishti280992/repro-mamba-icl-outliers-actual@d0daa81356fd508db339a949d418e462bb5ade6c/actual_outputs/convergence.csv', 'hf://datasets/Srishti280992/repro-mamba-icl-outliers-actual@d0daa81356fd508db339a949d418e462bb5ade6c/actual_outputs/fig2.csv', 'hf://datasets/Srishti280992/repro-mamba-icl-outliers-actual@d0daa81356fd508db339a949d418e462bb5ade6c/actual_outputs/mechanism.csv', 'hf://datasets/Srishti280992/repro-mamba-icl-outliers-actual@d0daa81356fd508db339a949d418e462bb5ade6c/actual_outputs/pa_threshold.csv', 'hf://datasets/Srishti280992/repro-mamba-icl-outliers-actual@d0daa81356fd508db339a949d418e462bb5ade6c/actual_outputs/table1.csv']

Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1837, in _prepare_split_single
                  writer.write_table(table)
                  ~~~~~~~~~~~~~~~~~~^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/arrow_writer.py", line 765, in write_table
                  self._write_table(pa_table, writer_batch_size=writer_batch_size)
                  ~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/arrow_writer.py", line 773, in _write_table
                  pa_table = table_cast(pa_table, self._schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2369, in table_cast
                  return cast_table_to_schema(table, schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2297, in cast_table_to_schema
                  raise CastError(
                  ...<3 lines>...
                  )
              datasets.table.CastError: Couldn't cast
              outlier_type: string
              arch: string
              alpha: double
              error: double
              -- schema metadata --
              pandas: '{"index_columns": [{"kind": "range", "name": null, "start": 0, "' + 732
              to
              {'arch': Value('string'), 'iter': Value('int64'), 'error': Value('float64')}
              because column names don't match
              
              During handling of the above exception, another exception occurred:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1369, in compute_config_parquet_and_info_response
                  parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
                                                                        ~~~~~~~~~~~~~~~~~~~~~~~~~^
                      builder, max_dataset_size_bytes=max_dataset_size_bytes
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  )
                  ^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 948, in stream_convert_to_parquet
                  builder._prepare_split(split_generator=splits_generators[split], file_format="parquet")
                  ~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1683, in _prepare_split
                  for job_id, done, content in self._prepare_split_single(
                                               ~~~~~~~~~~~~~~~~~~~~~~~~~~^
                      gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  ):
                  ^
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1839, in _prepare_split_single
                  raise DatasetGenerationCastError.from_cast_error(
                  ...<4 lines>...
                  )
              datasets.exceptions.DatasetGenerationCastError: An error occurred while generating the dataset
              
              All the data files must have the same columns, but at some point there are 2 new columns ({'outlier_type', 'alpha'}) and 1 missing columns ({'iter'}).
              
              This happened while the csv dataset builder was generating data using
              
              hf://datasets/Srishti280992/repro-mamba-icl-outliers-actual/actual_outputs/fig2.csv (at revision d0daa81356fd508db339a949d418e462bb5ade6c), ['hf://datasets/Srishti280992/repro-mamba-icl-outliers-actual@d0daa81356fd508db339a949d418e462bb5ade6c/actual_outputs/convergence.csv', 'hf://datasets/Srishti280992/repro-mamba-icl-outliers-actual@d0daa81356fd508db339a949d418e462bb5ade6c/actual_outputs/fig2.csv', 'hf://datasets/Srishti280992/repro-mamba-icl-outliers-actual@d0daa81356fd508db339a949d418e462bb5ade6c/actual_outputs/mechanism.csv', 'hf://datasets/Srishti280992/repro-mamba-icl-outliers-actual@d0daa81356fd508db339a949d418e462bb5ade6c/actual_outputs/pa_threshold.csv', 'hf://datasets/Srishti280992/repro-mamba-icl-outliers-actual@d0daa81356fd508db339a949d418e462bb5ade6c/actual_outputs/table1.csv']
              
              Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)

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.

arch
string
iter
int64
error
float64
Mamba
1
0.3295
Mamba
2
0.3315
Mamba
5
0.3345
Mamba
10
0.3295
Mamba
20
0.2915
Mamba
40
0.1045
Mamba
70
0.0275
Mamba
100
0.0105
Mamba
150
0.007
Mamba
200
0.006
Mamba
300
0.003
Mamba
500
0.0005
Mamba
800
0
Mamba
1,200
0
Mamba
1,800
0
Mamba
2,500
0
Mamba
4,000
0
LT
1
0.034
LT
2
0.0185
LT
5
0.04
LT
10
0.0265
LT
20
0.023
LT
40
0.0055
LT
70
0.005
LT
100
0.0005
LT
150
0
LT
200
0
LT
300
0
LT
500
0
LT
800
0
LT
1,200
0
LT
1,800
0
LT
2,500
0
LT
4,000
0
Mamba
null
0
Mamba
null
0
Mamba
null
0.0001
Mamba
null
0.0005
Mamba
null
0.001
Mamba
null
0.0034
Mamba
null
0.0157
Mamba
null
0.0754
Mamba
null
0.2897
Mamba
null
0.7594
LT
null
0
LT
null
0
LT
null
0
LT
null
0.0001
LT
null
0.0035
LT
null
0.5884
LT
null
0.998
LT
null
1
LT
null
1
LT
null
1
Mamba
null
0
Mamba
null
0
Mamba
null
0.0001
Mamba
null
0.0004
Mamba
null
0.0008
Mamba
null
0.0021
Mamba
null
0.0067
Mamba
null
0.0256
Mamba
null
0.0912
Mamba
null
0.2825
LT
null
0
LT
null
0
LT
null
0
LT
null
0.0001
LT
null
0.0005
LT
null
0.0062
LT
null
0.0369
LT
null
0.1128
LT
null
0.2292
LT
null
0.3647
Mamba
null
0
Mamba
null
0
Mamba
null
0
Mamba
null
0
Mamba
null
0.0001
Mamba
null
0.0003
Mamba
null
0.0016
Mamba
null
0.0047
Mamba
null
0.0246
Mamba
null
0.1264
LT
null
0
LT
null
0
LT
null
0
LT
null
0
LT
null
0.0001
LT
null
0.0013
LT
null
0.0137
LT
null
0.0695
LT
null
0.1843
LT
null
0.3502
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
End of preview.

Actual Mamba ICL Outlier Reproduction Artifacts

Measured reproduction artifacts for ICML 2026 paper C41aLahRXZ, "How Can Mamba Learn In Context with Outliers and Generalize Provably?".

The scripts implement the paper synthetic setup: Eq. (3), Definitions 1-2, hinge-loss SGD, one-layer Mamba and linear Transformer control. Outputs in actual_outputs/ are from 5 seeds, 4000 SGD iterations, batch 128, and 4000 prompts per evaluation point.

HF GPU evidence: https://huggingface.co/jobs/Srishti280992/6a630dc9db23d7a7ec1c9f09

Downloads last month
27