Datasets:

monsterDavid
/

affine-swe-terminal-sft-v1

Error code:   DatasetGenerationError
Exception:    CastError
Message:      Couldn't cast
built_at: string
inputs: struct<swe_sft: string, terminal_sft: string>
  child 0, swe_sft: string
  child 1, terminal_sft: string
balance: bool
seed: int64
input_counts: struct<swe: int64, terminal: int64>
  child 0, swe: int64
  child 1, terminal: int64
stats: struct<total: int64, by_environment: struct<SWE-INFINITE: int64, TERMINAL: int64>, by_environment_su (... 94 chars omitted)
  child 0, total: int64
  child 1, by_environment: struct<SWE-INFINITE: int64, TERMINAL: int64>
      child 0, SWE-INFINITE: int64
      child 1, TERMINAL: int64
  child 2, by_environment_subset: struct<SWE-INFINITE:default: int64, TERMINAL:stratified: int64, TERMINAL:random: int64>
      child 0, SWE-INFINITE:default: int64
      child 1, TERMINAL:stratified: int64
      child 2, TERMINAL:random: int64
output: string
messages: list<item: struct<role: string, content: string>>
  child 0, item: struct<role: string, content: string>
      child 0, role: string
      child 1, content: string
subset: string
environment: string
template_name: string
task_id: int64
score: double
to
{'task_id': Value('int64'), 'messages': List({'role': Value('string'), 'content': Value('string')}), 'score': Value('float64'), 'environment': Value('string'), 'subset': Value('string'), 'template_name': Value('string')}
because column names don't match
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1816, 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 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
              built_at: string
              inputs: struct<swe_sft: string, terminal_sft: string>
                child 0, swe_sft: string
                child 1, terminal_sft: string
              balance: bool
              seed: int64
              input_counts: struct<swe: int64, terminal: int64>
                child 0, swe: int64
                child 1, terminal: int64
              stats: struct<total: int64, by_environment: struct<SWE-INFINITE: int64, TERMINAL: int64>, by_environment_su (... 94 chars omitted)
                child 0, total: int64
                child 1, by_environment: struct<SWE-INFINITE: int64, TERMINAL: int64>
                    child 0, SWE-INFINITE: int64
                    child 1, TERMINAL: int64
                child 2, by_environment_subset: struct<SWE-INFINITE:default: int64, TERMINAL:stratified: int64, TERMINAL:random: int64>
                    child 0, SWE-INFINITE:default: int64
                    child 1, TERMINAL:stratified: int64
                    child 2, TERMINAL:random: int64
              output: string
              messages: list<item: struct<role: string, content: string>>
                child 0, item: struct<role: string, content: string>
                    child 0, role: string
                    child 1, content: string
              subset: string
              environment: string
              template_name: string
              task_id: int64
              score: double
              to
              {'task_id': Value('int64'), 'messages': List({'role': Value('string'), 'content': Value('string')}), 'score': Value('float64'), 'environment': Value('string'), 'subset': Value('string'), 'template_name': Value('string')}
              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 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 1869, 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_id int64	messages list	score float64	environment string	subset string	template_name string
20,253	[ { "role": "system", "content": "You are a helpful assistant that can interact multiple times with a computer shell to solve programming tasks.\nYour response must contain exactly ONE bash code block with ONE command (or commands connected with && or \|\|).\n\nInclude a THOUGHT section before your command wher...	1	SWE-INFINITE	null	null
85,000,002	[ { "role": "user", "content": "Translate a Gaussian Process RStan model (rbf) to PyStan 3.10.0.\n\n`/app/stan_r_script.R` — R script using RStan (DO NOT run, just read it)\n`/app/train_X.csv`, `/app/train_y.csv`, `/app/test_X.csv` — input data\n`/app/meta_public.json` — model hyperparameters (length_scale, s...	1	TERMINAL	stratified	RstanToPystan
19,997	[ { "role": "system", "content": "You are a helpful assistant that can interact multiple times with a computer shell to solve programming tasks.\nYour response must contain exactly ONE bash code block with ONE command (or commands connected with && or \|\|).\n\nInclude a THOUGHT section before your command wher...	1	SWE-INFINITE	null	null
20,605	[{"role":"system","content":"You are a helpful assistant that can interact multiple times with a com(...TRUNCATED)	1	SWE-INFINITE	null	null
20,599	[{"role":"system","content":"You are a helpful assistant that can interact multiple times with a com(...TRUNCATED)	1	SWE-INFINITE	null	null
20,025	[{"role":"system","content":"You are a helpful assistant that can interact multiple times with a com(...TRUNCATED)	1	SWE-INFINITE	null	null
28,538,251	[{"role":"user","content":"A security audit flagged the SSH daemon configuration as non-compliant.\n(...TRUNCATED)	1	TERMINAL	random	SshdHardening
20,171	[{"role":"system","content":"You are a helpful assistant that can interact multiple times with a com(...TRUNCATED)	1	SWE-INFINITE	null	null
4,216,175	[{"role":"user","content":"HTTP traffic is being rejected — the web server appears to be down.\n\n(...TRUNCATED)	1	TERMINAL	random	NginxConfigFix
35,575,298	[{"role":"user","content":"You have a rotation spec at `/root/rotation_spec.txt` that defines how lo(...TRUNCATED)	1	TERMINAL	random	LogRotationScript

End of preview.

Affine SWE + TERMINAL SFT Pass Datasets

Champion eval pass trajectories for SN120 training.

Files

File	Description	Records
`swe_terminal_sft_pass.jsonl`	Full mix (all SWE + all TERMINAL passes)	847
`swe_terminal_sft_pass.jsonl.gz`	Gzip of full mix	847
`balanced/swe_terminal_sft_pass.jsonl`	1:1 balanced mix (301 SWE + 301 TERMINAL)	602
`balanced/swe_terminal_sft_pass.jsonl.gz`	Gzip of balanced mix	602

Schema

Each line is JSON with: task_id, environment, messages, score (+ subset, template_name for TERMINAL).

Sources

SWE-INFINITE: 500-task champion eval, 301 passes
TERMINAL: random 400 + stratified 396 champion eval, 546 passes

Downloads last month: 47

Total file size:

59.2 MB