Hugging Face's logo

Datasets:
rl-rag-2
/
oss-high-bc-synthetic-v2-8rollouts

Tags:
deep-research
browsecomp
trajectories
License:
Dataset card Files
xet
 Community
Dataset Viewer
Duplicate
The dataset viewer is not available for this split.
Cannot load the dataset split (in streaming mode) to extract the first rows.
Error code:   StreamingRowsError
Exception:    CastError
Message:      Couldn't cast
question: string
answer: string
qid: string
reference_answer: string
traj_idx: int64
request_qid: string
messages: list<item: struct<role: string, name: string, content: list<item: struct<model_identity: string, rea (... 849 chars omitted)
  child 0, item: struct<role: string, name: string, content: list<item: struct<model_identity: string, reasoning_effo (... 837 chars omitted)
      child 0, role: string
      child 1, name: string
      child 2, content: list<item: struct<model_identity: string, reasoning_effort: string, conversation_start_date: timesta (... 734 chars omitted)
          child 0, item: struct<model_identity: string, reasoning_effort: string, conversation_start_date: timestamp[s], know (... 722 chars omitted)
              child 0, model_identity: string
              child 1, reasoning_effort: string
              child 2, conversation_start_date: timestamp[s]
              child 3, knowledge_cutoff: string
              child 4, channel_config: struct<valid_channels: list<item: string>, channel_required: bool>
                  child 0, valid_channels: list<item: string>
                      child 0, item: string
                  child 1, channel_required: bool
              child 5, tools: struct<browser: struct<name: string, description: string, tools: list<item: struct<name: string, des (... 480 chars omitted)
                  child 0, browser: struct<name: string, description: string, tools: list<item: struct<name: string, description: string (... 4
...
 6, num_lines: struct<type: string, default: int64>
                                          child 0, type: string
                                          child 1, default: int64
                                      child 7, view_source: struct<type: string, default: bool>
                                          child 0, type: string
                                          child 1, default: bool
                                      child 8, pattern: struct<type: string>
                                          child 0, type: string
                                  child 2, required: list<item: string>
                                      child 0, item: string
              child 6, type: string
              child 7, text: string
      child 3, channel: string
      child 4, recipient: string
      child 5, content_type: string
searched_urls: list<item: string>
  child 0, item: string
search_results: list<item: struct<query: string, rank: int64, title: string, url: string, summary: string>>
  child 0, item: struct<query: string, rank: int64, title: string, url: string, summary: string>
      child 0, query: string
      child 1, rank: int64
      child 2, title: string
      child 3, url: string
      child 4, summary: string
latency_s: double
error: null
attempts: int64
status: string
reasoning_effort: string
world_size: int64
num_trajectories: int64
total_trajectories: int64
assigned_questions: int64
rank: int64
num_questions: int64
split: string
dataset: string
to
{'dataset': Value('string'), 'split': Value('string'), 'num_questions': Value('int64'), 'num_trajectories': Value('int64'), 'total_trajectories': Value('int64'), 'assigned_questions': Value('int64'), 'rank': Value('int64'), 'world_size': Value('int64'), 'reasoning_effort': Value('string')}
because column names don't match
Traceback:    Traceback (most recent call last):
                File "/src/services/worker/src/worker/utils.py", line 99, in get_rows_or_raise
                  return get_rows(
                         ^^^^^^^^^
                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 77, in get_rows
                  rows_plus_one = list(itertools.islice(ds, rows_max_number + 1))
                                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2690, in __iter__
                  for key, example in ex_iterable:
                                      ^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2227, in __iter__
                  for key, pa_table in self._iter_arrow():
                                       ^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2251, in _iter_arrow
                  for key, pa_table in self.ex_iterable._iter_arrow():
                                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 494, in _iter_arrow
                  for key, pa_table in iterator:
                                       ^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 384, in _iter_arrow
                  for key, pa_table in self.generate_tables_fn(**gen_kwags):
                                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 295, in _generate_tables
                  self._cast_table(pa_table, json_field_paths=json_field_paths),
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 128, in _cast_table
                  pa_table = table_cast(pa_table, self.info.features.arrow_schema)
                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2281, in table_cast
                  return cast_table_to_schema(table, schema)
                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2227, in cast_table_to_schema
                  raise CastError(
              datasets.table.CastError: Couldn't cast
              question: string
              answer: string
              qid: string
              reference_answer: string
              traj_idx: int64
              request_qid: string
              messages: list<item: struct<role: string, name: string, content: list<item: struct<model_identity: string, rea (... 849 chars omitted)
                child 0, item: struct<role: string, name: string, content: list<item: struct<model_identity: string, reasoning_effo (... 837 chars omitted)
                    child 0, role: string
                    child 1, name: string
                    child 2, content: list<item: struct<model_identity: string, reasoning_effort: string, conversation_start_date: timesta (... 734 chars omitted)
                        child 0, item: struct<model_identity: string, reasoning_effort: string, conversation_start_date: timestamp[s], know (... 722 chars omitted)
                            child 0, model_identity: string
                            child 1, reasoning_effort: string
                            child 2, conversation_start_date: timestamp[s]
                            child 3, knowledge_cutoff: string
                            child 4, channel_config: struct<valid_channels: list<item: string>, channel_required: bool>
                                child 0, valid_channels: list<item: string>
                                    child 0, item: string
                                child 1, channel_required: bool
                            child 5, tools: struct<browser: struct<name: string, description: string, tools: list<item: struct<name: string, des (... 480 chars omitted)
                                child 0, browser: struct<name: string, description: string, tools: list<item: struct<name: string, description: string (... 4
              ...
               6, num_lines: struct<type: string, default: int64>
                                                        child 0, type: string
                                                        child 1, default: int64
                                                    child 7, view_source: struct<type: string, default: bool>
                                                        child 0, type: string
                                                        child 1, default: bool
                                                    child 8, pattern: struct<type: string>
                                                        child 0, type: string
                                                child 2, required: list<item: string>
                                                    child 0, item: string
                            child 6, type: string
                            child 7, text: string
                    child 3, channel: string
                    child 4, recipient: string
                    child 5, content_type: string
              searched_urls: list<item: string>
                child 0, item: string
              search_results: list<item: struct<query: string, rank: int64, title: string, url: string, summary: string>>
                child 0, item: struct<query: string, rank: int64, title: string, url: string, summary: string>
                    child 0, query: string
                    child 1, rank: int64
                    child 2, title: string
                    child 3, url: string
                    child 4, summary: string
              latency_s: double
              error: null
              attempts: int64
              status: string
              reasoning_effort: string
              world_size: int64
              num_trajectories: int64
              total_trajectories: int64
              assigned_questions: int64
              rank: int64
              num_questions: int64
              split: string
              dataset: string
              to
              {'dataset': Value('string'), 'split': Value('string'), 'num_questions': Value('int64'), 'num_trajectories': Value('int64'), 'total_trajectories': Value('int64'), 'assigned_questions': Value('int64'), 'rank': Value('int64'), 'world_size': Value('int64'), 'reasoning_effort': Value('string')}
              because column names don't match

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.

OSS High-Effort Trajectories: bc_synthetic_v2

Model

  • Model: gpt-oss-120b (internal)
  • Reasoning effort: high
  • Engine: OSS (Harmony format with browser tools)

Source Data

  • Dataset: rl-rag/bc_synthetic_v_2
  • Split: normal
  • Questions: 1,996
  • Rollouts per question: 8
  • Total trajectories: 15,967

Format

Each line in trajectories.jsonl is a JSON object with:

  • qid: question ID
  • question: the input question
  • answer: reference answer from source dataset
  • traj_idx: rollout index (0-7)
  • messages: full Harmony-format conversation including:
    • System message (with reasoning_effort, tool config)
    • Developer message (browser tool instructions)
    • User message
    • All assistant analysis/thinking messages
    • All tool calls (browser.search, browser.open, browser.find) with arguments
    • All tool responses with full content
    • Final answer (channel=final)
  • searched_urls: list of unique full URLs returned by browser search calls
  • search_results: list of full search result records, each with:
    • query: the search query
    • rank: result rank
    • title: result title
    • url: full URL
    • summary: snippet/summary
  • latency_s: generation time in seconds
  • status: "success"

Tools

  • browser.search: web search via Serper API
  • browser.open: fetch and render web pages
  • browser.find: find text patterns in opened pages

Citation Format

Final answers use Harmony citation format: 【cursor†Lstart-Lend】
Downloads last month
20
Total file size:
4.42 GB