Datasets:

JoyboyXoXo
/

agentic-failure-attribution-traces

Error code:   StreamingRowsError
Exception:    TypeError
Message:      Couldn't cast array of type
struct<2: struct<taskAlignment: double, progress: double, repetition: double, stuck: double, needsVerification: double, prematureCompletion: double, unexpectedDirection: double, phase: string>, 5: struct<taskAlignment: double, progress: double, repetition: double, stuck: double, needsVerification: double, prematureCompletion: double, unexpectedDirection: double, phase: string>, 7: struct<taskAlignment: double, progress: double, repetition: double, stuck: double, needsVerification: double, prematureCompletion: double, unexpectedDirection: double, phase: string>, 9: struct<taskAlignment: double, progress: double, repetition: double, stuck: double, needsVerification: double, prematureCompletion: double, unexpectedDirection: double, phase: string>, 11: struct<taskAlignment: double, progress: double, repetition: double, stuck: double, needsVerification: double, prematureCompletion: double, unexpectedDirection: double, phase: string>, 13: struct<taskAlignment: double, progress: double, repetition: double, stuck: double, needsVerification: double, prematureCompletion: double, unexpectedDirection: double, phase: string>>
to
{'2': {'taskAlignment': Value('float64'), 'progress': Value('float64'), 'repetition': Value('float64'), 'stuck': Value('float64'), 'needsVerification': Value('float64'), 'prematureCompletion': Value('float64'), 'unexpectedDirection': Value('float64'), 'phase': Value('string')}, '5': {'taskAlignment': Value('float64'), 'progress': Value('float64'), 'repetition': Value('float64'), 'stuck': Value('float64'), 'needsVerification': Value('float64'), 'prematureCompletion': Value('float64'), 'unexpectedDirection': Value('float64'), 'phase': Value('string')}, '8': {'taskAlignment': Value('float64'), 'progress': Value('float64'), 'repetition': Value('float64'), 'stuck': Value('float64'), 'needsVerification': Value('float64'), 'prematureCompletion': Value('float64'), 'unexpectedDirection': Value('float64'), 'phase': Value('string')}, '11': {'taskAlignment': Value('float64'), 'progress': Value('float64'), 'repetition': Value('float64'), 'stuck': Value('float64'), 'needsVerification': Value('float64'), 'prematureCompletion': Value('float64'), 'unexpectedDirection': Value('float64'), 'phase': Value('string')}, '13': {'taskAlignment': Value('float64'), 'progress': Value('float64'), 'repetition': Value('float64'), 'stuck': Value('float64'), 'needsVerification': Value('float64'), 'prematureCompletion': Value('float64'), 'unexpectedDirection': Value('float64'), 'phase': Value('string')}, '16': {'taskAlignment': Value('float64'), 'progress': Value('float64'), 'repetition': Value('float64'), 'stuck': Value('float64'), 'needsVerification': Value('float64'), 'prematureCompletion': Value('float64'), 'unexpectedDirection': Value('float64'), 'phase': Value('string')}}
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 2161, 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<2: struct<taskAlignment: double, progress: double, repetition: double, stuck: double, needsVerification: double, prematureCompletion: double, unexpectedDirection: double, phase: string>, 5: struct<taskAlignment: double, progress: double, repetition: double, stuck: double, needsVerification: double, prematureCompletion: double, unexpectedDirection: double, phase: string>, 7: struct<taskAlignment: double, progress: double, repetition: double, stuck: double, needsVerification: double, prematureCompletion: double, unexpectedDirection: double, phase: string>, 9: struct<taskAlignment: double, progress: double, repetition: double, stuck: double, needsVerification: double, prematureCompletion: double, unexpectedDirection: double, phase: string>, 11: struct<taskAlignment: double, progress: double, repetition: double, stuck: double, needsVerification: double, prematureCompletion: double, unexpectedDirection: double, phase: string>, 13: struct<taskAlignment: double, progress: double, repetition: double, stuck: double, needsVerification: double, prematureCompletion: double, unexpectedDirection: double, phase: string>>
              to
              {'2': {'taskAlignment': Value('float64'), 'progress': Value('float64'), 'repetition': Value('float64'), 'stuck': Value('float64'), 'needsVerification': Value('float64'), 'prematureCompletion': Value('float64'), 'unexpectedDirection': Value('float64'), 'phase': Value('string')}, '5': {'taskAlignment': Value('float64'), 'progress': Value('float64'), 'repetition': Value('float64'), 'stuck': Value('float64'), 'needsVerification': Value('float64'), 'prematureCompletion': Value('float64'), 'unexpectedDirection': Value('float64'), 'phase': Value('string')}, '8': {'taskAlignment': Value('float64'), 'progress': Value('float64'), 'repetition': Value('float64'), 'stuck': Value('float64'), 'needsVerification': Value('float64'), 'prematureCompletion': Value('float64'), 'unexpectedDirection': Value('float64'), 'phase': Value('string')}, '11': {'taskAlignment': Value('float64'), 'progress': Value('float64'), 'repetition': Value('float64'), 'stuck': Value('float64'), 'needsVerification': Value('float64'), 'prematureCompletion': Value('float64'), 'unexpectedDirection': Value('float64'), 'phase': Value('string')}, '13': {'taskAlignment': Value('float64'), 'progress': Value('float64'), 'repetition': Value('float64'), 'stuck': Value('float64'), 'needsVerification': Value('float64'), 'prematureCompletion': Value('float64'), 'unexpectedDirection': Value('float64'), 'phase': Value('string')}, '16': {'taskAlignment': Value('float64'), 'progress': Value('float64'), 'repetition': Value('float64'), 'stuck': Value('float64'), 'needsVerification': Value('float64'), 'prematureCompletion': Value('float64'), 'unexpectedDirection': Value('float64'), 'phase': Value('string')}}

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.

Agentic Failure Attribution: τ-bench traces with Jev annotations

The conversation data behind the study Agentic Failure Attribution: what agent traces do and don't tell you about failure.

7,750 conversations from the τ-bench (tau2-bench) banking_knowledge domain: 20 models, 97 tasks, 4 trials each (10 runs with no messages are left out).
Outcome labels re-scored with the tau2-bench v1.0.1 grading code.
Jev annotations for 1,550 conversations (Kimi K3, Claude Opus 5, GPT-5.5, Qwen 3.8 Max): 9,257 step judgments and 8,913 user-message pushback checks from TypeSafe Jev 1.14.

Files

One JSON file per run: runs/<run_id>.json.

Field	Meaning
`run_id`, `model`, `task_id`, `trial`	Which run this is
`reward`	1 = passed, 0 = failed (re-scored)
`decisive`	`{idx, kind}`: the agent's first extra or wrong write action (`unexpected_write`, `wrong_write`), or the last message (`end_of_run`) if the expected write never happened
`messages`	`[{idx, role, turn, text, tool, args}]`. A message with several tool calls has one row per call, with the same `idx`. Tool results and tool arguments are cut to 1,500 characters
`judgments`	Jev models only. Keyed by message `idx`: probabilities for `taskAlignment`, `progress`, `repetition`, `stuck`, `needsVerification`, `prematureCompletion`, `unexpectedDirection`, and the `phase` choice. Only the first 6 agent steps after the greeting are judged
`pushback`	Jev models only. Keyed by user message `idx`: probabilities for `corrects`, `repeats`, `complains`, `asksHuman`, `saysDone`, and `pushback` = the max of the first 4

Limits

The customer is an LLM user simulator (gpt-5.2) that follows a task script, not a real customer.
The Jev scores are model judgments. The pushback check was validated on 60 messages labelled by another LLM, not by humans: recall 1.00, precision 0.48 at threshold 0.5.

Sources and license

Conversations, tasks and grading: τ-bench / tau2-bench, Sierra, MIT license.
Judgments: TypeSafe Jev 1.14.
This dataset: MIT.

Downloads last month: -

Total file size:

382 MB