Datasets:

raihan-js
/

kotoba-tutor-eval

License:

Dataset Preview

Split (2)

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
n: int64
method: string
false_correction: double
per_type: struct<particle: struct<recall: double, n: int64>, conjugation: struct<recall: double, n: int64>, po (... 85 chars omitted)
  child 0, particle: struct<recall: double, n: int64>
      child 0, recall: double
      child 1, n: int64
  child 1, conjugation: struct<recall: double, n: int64>
      child 0, recall: double
      child 1, n: int64
  child 2, politeness: struct<recall: double, n: int64>
      child 0, recall: double
      child 1, n: int64
  child 3, choice: struct<recall: double, n: int64>
      child 0, recall: double
      child 1, n: int64
strict_correction: double
confusion: struct<choice>choice: int64, choice>none: int64, choice>particle: int64, choice>politeness: int64, c (... 342 chars omitted)
  child 0, choice>choice: int64
  child 1, choice>none: int64
  child 2, choice>particle: int64
  child 3, choice>politeness: int64
  child 4, conjugation>choice: int64
  child 5, conjugation>none: int64
  child 6, none>choice: int64
  child 7, none>conjugation: int64
  child 8, none>none: int64
  child 9, none>particle: int64
  child 10, none>politeness: int64
  child 11, particle>choice: int64
  child 12, particle>none: int64
  child 13, particle>particle: int64
  child 14, politeness>choice: int64
  child 15, politeness>conjugation: int64
  child 16, politeness>none: int64
  child 17, politeness>politeness: int64
steps: int64
final_loss: double
rows: int64
epochs: int64
secs: int64
to
{'epochs': Value('int64'), 'rows': Value('int64'), 'steps': Value('int64'), 'final_loss': Value('float64'), 'secs': 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
              n: int64
              method: string
              false_correction: double
              per_type: struct<particle: struct<recall: double, n: int64>, conjugation: struct<recall: double, n: int64>, po (... 85 chars omitted)
                child 0, particle: struct<recall: double, n: int64>
                    child 0, recall: double
                    child 1, n: int64
                child 1, conjugation: struct<recall: double, n: int64>
                    child 0, recall: double
                    child 1, n: int64
                child 2, politeness: struct<recall: double, n: int64>
                    child 0, recall: double
                    child 1, n: int64
                child 3, choice: struct<recall: double, n: int64>
                    child 0, recall: double
                    child 1, n: int64
              strict_correction: double
              confusion: struct<choice>choice: int64, choice>none: int64, choice>particle: int64, choice>politeness: int64, c (... 342 chars omitted)
                child 0, choice>choice: int64
                child 1, choice>none: int64
                child 2, choice>particle: int64
                child 3, choice>politeness: int64
                child 4, conjugation>choice: int64
                child 5, conjugation>none: int64
                child 6, none>choice: int64
                child 7, none>conjugation: int64
                child 8, none>none: int64
                child 9, none>particle: int64
                child 10, none>politeness: int64
                child 11, particle>choice: int64
                child 12, particle>none: int64
                child 13, particle>particle: int64
                child 14, politeness>choice: int64
                child 15, politeness>conjugation: int64
                child 16, politeness>none: int64
                child 17, politeness>politeness: int64
              steps: int64
              final_loss: double
              rows: int64
              epochs: int64
              secs: int64
              to
              {'epochs': Value('int64'), 'rows': Value('int64'), 'steps': Value('int64'), 'final_loss': Value('float64'), 'secs': 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 1879, 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.

epochs int64	rows int64	steps int64	final_loss float64	secs int64
2	50,000	3,124	0.6502	10,180
2	60,050	2,076	1.665	6,135

kotoba-tutor-eval

Evaluation artifacts for kotoba-tutor (voice-first Japanese tutor agent in LangGraph, all-local pipeline).

m1_rows.jsonl: 300 correction items — 150 correct + 150 with one rule-injected error (particle/conjugation/politeness/choice). Correct sentences: Tatoeba jpn_sentences (CC-BY 2.0 FR, (c) Tatoeba contributors, used under the licence). Gold = Tatoeba original; injection tags exact by construction (no fluent checker — disclosed).
m1.json: dataset stats. m3.json + m3_rows.jsonl: memory-vs-memoryless eval (frozen prompt, see prereg_m3.md). vram.json: M0 VRAM spike.

Headline M3: the Swallow-8B analyse node predicts "particle" on 285/300 rows and never the other three types; memory reminder has no detectable effect. v1.1 (v11.json, rates only): same frozen prompt on 1,288 TEC-JL human-gold rows — false-correction 0.95 [0.93, 0.97], collapse replicates. TEC-JL row texts are research-only and never redistributed (not in this dataset).

Downloads last month: 45

Total file size:

335 kB