Datasets:

dCoder30
/

gamedoctor-sft-data

Tasks:

Text Generation

Tags:

agentic-sft

tool-use

code-generation

License:

Dataset card Data Studio Files Files and versions

xet

Community

Dataset Preview

Duplicate

Split (1)

train

The full dataset viewer is not available (click to read why). Only showing a preview of the rows.

The dataset generation failed

Error code:   DatasetGenerationError
Exception:    CastError
Message:      Couldn't cast
task: string
archetype: string
model: string
n_chars: int64
engine: string
source: string
messages: list<item: string>
  child 0, item: string
session_terminated: bool
n_sessions_dropped: int64
success: bool
todo_calls_dropped: int64
stopped_by: null
incomplete_calls_dropped: int64
n_sessions_in_file: int64
wall_ms: int64
n_tokens: int64
game_dir: string
n_tool_calls: int64
to
{'task': Value('string'), 'model': Value('string'), 'n_tool_calls': Value('int64'), 'n_sessions_in_file': Value('int64'), 'n_sessions_dropped': Value('int64'), 'todo_calls_dropped': Value('int64'), 'incomplete_calls_dropped': Value('int64'), 'success': Value('bool'), 'session_terminated': Value('bool'), 'stopped_by': Value('null'), 'wall_ms': Value('int64'), 'messages': List(Json(decode=True)), 'game_dir': Value('string'), 'n_tokens': Value('int64')}
because column names don't match
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1827, in _prepare_split_single
                  for key, table in generator:
                                    ^^^^^^^^^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 613, in wrapped
                  for item in generator(*args, **kwargs):
                              ~~~~~~~~~^^^^^^^^^^^^^^^^^
                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 2306, in cast_table_to_schema
                  raise CastError(
                  ...<3 lines>...
                  )
              datasets.table.CastError: Couldn't cast
              task: string
              archetype: string
              model: string
              n_chars: int64
              engine: string
              source: string
              messages: list<item: string>
                child 0, item: string
              session_terminated: bool
              n_sessions_dropped: int64
              success: bool
              todo_calls_dropped: int64
              stopped_by: null
              incomplete_calls_dropped: int64
              n_sessions_in_file: int64
              wall_ms: int64
              n_tokens: int64
              game_dir: string
              n_tool_calls: int64
              to
              {'task': Value('string'), 'model': Value('string'), 'n_tool_calls': Value('int64'), 'n_sessions_in_file': Value('int64'), 'n_sessions_dropped': Value('int64'), 'todo_calls_dropped': Value('int64'), 'incomplete_calls_dropped': Value('int64'), 'success': Value('bool'), 'session_terminated': Value('bool'), 'stopped_by': Value('null'), 'wall_ms': Value('int64'), 'messages': List(Json(decode=True)), 'game_dir': Value('string'), 'n_tokens': Value('int64')}
              because column names don't match
              
              The above exception was the direct cause of the following exception:
              
              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 1694, 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 1880, in _prepare_split_single
                  raise DatasetGenerationError("An error occurred while generating the dataset") from e
              datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset

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.

task string	model string	n_tool_calls int64	n_sessions_in_file int64	n_sessions_dropped int64	todo_calls_dropped int64	success bool	session_terminated bool	stopped_by null	wall_ms int64	messages list	game_dir string	n_tokens int64
bounce-arcade-D3-001	gpt-5.5	80	2	1	15	true	true	null	1,876,332	[ { "role": "user", "content": "Please follow the Top-Down Archetype workflow.\n\nBuild a small polished Phaser bounce game called \"Lantern Rebound\".\nTheme: Feudal-Japan ninja courtyard under ink-wash skies, warm paper lanterns, cherry blossoms, roof shadows, and crisp brush-stroke impacts.\nSignature twis...	/tmp/gamedoctor_basegen_1500/bounce-arcade-D3-001	32,893
bounce-arcade-D3-002	gpt-5.5	85	3	2	14	true	true	null	1,536,580	[{"role":"user","content":"Please follow the Top-Down Archetype workflow.\n\nBuild a small polished (...TRUNCATED)	/tmp/gamedoctor_basegen_1500/bounce-arcade-D3-002	42,273
bounce-arcade-D3-003	gpt-5.5	90	2	1	14	true	true	null	1,811,944	[{"role":"user","content":"Please follow the Top-Down Archetype workflow.\n\nBuild a small polished (...TRUNCATED)	/tmp/gamedoctor_basegen_1500/bounce-arcade-D3-003	49,025
bounce-arcade-D3-004	gpt-5.5	78	2	1	14	true	true	null	1,510,938	[{"role":"user","content":"Please follow the Top-Down Archetype workflow.\n\nBuild a small polished (...TRUNCATED)	/tmp/gamedoctor_basegen_1500/bounce-arcade-D3-004	36,494
bounce-arcade-D3-005	gpt-5.5	101	1	0	15	true	true	null	1,807,357	[{"role":"user","content":"Please follow the Top-Down Archetype workflow.\n\nBuild a small polished (...TRUNCATED)	/tmp/gamedoctor_basegen_1500/bounce-arcade-D3-005	45,837
bounce-arcade-D3-006	gpt-5.5	114	3	2	28	true	true	null	2,499,697	[{"role":"user","content":"Please follow the Top-Down Archetype workflow.\n\nBuild a small polished (...TRUNCATED)	/tmp/gamedoctor_basegen_1500/bounce-arcade-D3-006	63,128
bounce-arcade-D3-007	gpt-5.5	106	1	0	14	true	true	null	1,592,888	[{"role":"user","content":"Please follow the Top-Down Archetype workflow.\n\nBuild a small polished (...TRUNCATED)	/tmp/gamedoctor_basegen_1500/bounce-arcade-D3-007	41,011
bounce-arcade-D3-008	gpt-5.5	103	1	0	12	true	true	null	1,625,838	[{"role":"user","content":"Please follow the Top-Down Archetype workflow.\n\nBuild a small polished (...TRUNCATED)	/tmp/gamedoctor_basegen_1500/bounce-arcade-D3-008	47,756
bounce-arcade-D3-009	gpt-5.5	91	1	0	13	true	true	null	1,855,157	[{"role":"user","content":"Please follow the Top-Down Archetype workflow.\n\nBuild a small polished (...TRUNCATED)	/tmp/gamedoctor_basegen_1500/bounce-arcade-D3-009	47,265
bounce-arcade-D3-010	gpt-5.5	92	1	0	19	true	true	null	2,219,620	[{"role":"user","content":"Please follow the Top-Down Archetype workflow.\n\nBuild a small polished (...TRUNCATED)	/tmp/gamedoctor_basegen_1500/bounce-arcade-D3-010	47,005

End of preview.

GameDoctor agentic-SFT corpora

Training corpora for a three-stage agentic-SFT ablation on Qwen3.8-27B. Each row is one multi-turn agent trajectory in messages form, with tool calls and tool results, distilled from a stronger teacher building browser games.

Files

file	rows	what it is
`basegen_agentic_128k_v3.jsonl`	409	gen — build a whole game from a brief
`plan.jsonl`	454	plan — write the plan before building
`repair.jsonl`	145	repair — fix a broken game
`repair_negatives.jsonl`	96	repair negatives
`stage2_gen_plan.jsonl`	863	stage-2 v1 = gen + plan
`stage2_gen_plan_rw.jsonl`	525	stage-2 reweighted = 409 gen + 116 plan
`stage3_gen_plan_repair.jsonl`	1,008	stage-3 v1 = gen + plan + repair
`stage3_gen_plan_repair_rw.jsonl`	572	stage-3 reweighted = 409 + 116 + 47

v1 vs reweighted

Same examples, different frequency. Plan and repair are subsampled to 25%, stratified by archetype and task, seed 42. Step share in stage 2:

	v1	reweighted
gen	409/863 = 47.4%	409/525 = 77.9%
plan	454/863 = 52.6%	116/525 = 22.1%

Row schema

task, model, n_tool_calls, n_sessions_in_file, n_sessions_dropped,
todo_calls_dropped, incomplete_calls_dropped, success, session_terminated,
stopped_by, wall_ms, messages, game_dir, n_tokens

messages is the trained field. Everything else is provenance kept so any row can be traced back to the trajectory it came from.

How trajectories became rows

The transform (traj_to_sft_v2.py in the code repo) is not a flatten — several rules exist because the naive version teaches the wrong thing:

Session selection. A trajectory file holds several sessions. Kept: the last session that made tool calls, and only if dist/index.html exists. Success is judged by the artefact, not by a success flag.
Pair tool results on id, never on position. A result can arrive after the next call has started.
Consecutive tool calls collapse into ONE assistant message with multiple tool_calls, matching how the model is served.
Unanswered tool calls are dropped. Keeping them teaches a call answered by silence.
Oversized results are truncated head+tail (1,200 / 800 chars), shell output capped at 2,000. Late todo_write calls are dropped and acknowledged with a placeholder.
Trajectories under 10 KiB are discarded.

Only assistant turns are supervised.

Two hypotheses that were tested and rejected

Worth recording so nobody re-derives them:

Leaked absolute paths. Suspected that trajectories taught host-specific paths. Probed both models directly: both emit {"path": "src"}. No leak.
Past-tense plan targets. Suspected that plan rows described work already done, from selecting the wrong session. Measured: single-session 95.6% past-tense vs multi-session 97.4% — session choice was never the cause. A re-extraction that demanded forward-looking plans salvaged 7 rows of 454, which would have destroyed the corpus rather than fixed it.

Neither was changed. Reweighting is the only difference between v1 and rw.

Provenance and licence

Trajectories are model-generated (teacher outputs), not scraped or human-authored. Redistribution terms follow whatever governs the teacher model's outputs and the originating project; treat as internal unless cleared otherwise.

Pipeline code, reports and scores: WenyiWU0111/OpenGame-reproduce · gamedoctor_sft/

Companion repos: adapters · training data · eval results

Downloads last month: 98

Total file size:

862 MB