Datasets:
The dataset viewer is not available for this split.
Error code: StreamingRowsError
Exception: TypeError
Message: Couldn't cast array of type
struct<mandatory_recall: double, critical_recall: double, n_mandatory: int64, n_critical: int64, missing_mandatory: list<item: null>, missing_critical: list<item: null>>
to
{}
Traceback: Traceback (most recent call last):
File "/src/services/worker/src/worker/utils.py", line 147, 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 127, 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 483, 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 2840, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2373, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2398, 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 536, in _iter_arrow
for key, pa_table in iterator:
^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 419, 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 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 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 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 2068, in cast_array_to_feature
_c(array.field(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 2068, in cast_array_to_feature
_c(array.field(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 2158, in cast_array_to_feature
raise TypeError(f"Couldn't cast array of type\n{_short_str(array.type)}\nto\n{_short_str(feature)}")
TypeError: Couldn't cast array of type
struct<mandatory_recall: double, critical_recall: double, n_mandatory: int64, n_critical: int64, missing_mandatory: list<item: null>, missing_critical: list<item: null>>
to
{}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.
Viscer corpus v1 — synthetic coding-agent sessions with verification ledgers
Training and evaluation data for Viscer-4B, a model that compresses a coding-agent session into structured JSON state.
All sessions are synthetic, generated chapter-by-chapter by a larger hosted model. No real user data, no proprietary code, no personal information.
Contents
| Path | Contents |
|---|---|
sessions/ |
294 sessions across 14 domains × 10 session shapes (140/140 combinations populated), 112 S / 98 M / 70 L / 14 XL by length; XL cases are real 140K-token sessions |
targets/ |
383 verified compression targets: 262 positive + 88 negative/fallback, each with a quality record |
manifests/ |
ledger pruning log, diversity audit, holdout split assignment |
Each session file carries its transcript, a ground-truth fact ledger (mandatory / optional / superseded / forbidden-to-infer), and a transcript hash. Each target file carries the gold JSON state plus verification results.
How the targets were verified
Every target is checked programmatically against its ledger: JSON-schema conformance, ≥90% recall of critical literals, all hard constraints present, zero false completions, and budget treated as a soft criterion. 48 generated targets were rejected by these checks; the rejection taxonomy is reported in the accompanying technical report.
No independent human audit has been performed — verification is programmatic only. This is disclosed rather than implied.
Domains
Python debugging · C++ build systems · Unreal Engine · Git operations · shell/filesystem/MCP-heavy agent work · mixed long sessions · Rust/Go · TypeScript frontend · SQL/databases · Kubernetes/DevOps · security patching · ML pipelines · embedded C · data pipelines
Session shapes
debug-fix · feature-add · refactor · migration/upgrade · incident response · performance tuning · test authoring · code review · release prep · multi-goal
Intended use
Fine-tuning or evaluating context-compression models. The negative/fallback
targets (schema field handling = passthrough or refused) exist so that
models learn when not to compress, which is where reliability usually fails.
License
Apache-2.0, matching the base model lineage.
Citation
See CITATION.cff in the model repository.
- Downloads last month
- 26