kelsbeans/qwen3-1.7b-digestive-coach-n1102
2B • Updated • 337
Datasets:
Dataset Preview
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
scenario_id: string
family_id: string
user_message: string
context: struct<bleeding_reported: bool, vomiting_persistent: bool, fever_reported: bool, weight_loss_lb: int (... 160 chars omitted)
child 0, bleeding_reported: bool
child 1, vomiting_persistent: bool
child 2, fever_reported: bool
child 3, weight_loss_lb: int64
child 4, weight_loss_intentional: bool
child 5, inability_to_pass_gas: bool
child 6, severe_progressive_pain: bool
child 7, melena: bool
child 8, abdominal_pain: bool
child 9, iron_deficiency_anemia: bool
gold: struct<care_level: string, red_flags: list<item: string>, tcm_pattern: string, arbitration: list<ite (... 177 chars omitted)
child 0, care_level: string
child 1, red_flags: list<item: string>
child 0, item: string
child 2, tcm_pattern: string
child 3, arbitration: list<item: struct<intervention: string, outcome: string, relationship: string, evidence_level: strin (... 3 chars omitted)
child 0, item: struct<intervention: string, outcome: string, relationship: string, evidence_level: string>
child 0, intervention: string
child 1, outcome: string
child 2, relationship: string
child 3, evidence_level: string
child 4, forbidden: list<item: struct<intervention: string, as: string, reason: string>>
child 0, item: struct<intervention: string, as: string, reason: string>
child 0, intervention: string
child 1, as: string
child 2, reason: string
cl
...
ities: list<item: string>
child 0, item: string
child 3, arbitration: list<item: struct<intervention: string, outcome: string, relationship: string, evidence_level: strin (... 49 chars omitted)
child 0, item: struct<intervention: string, outcome: string, relationship: string, evidence_level: string, risk: st (... 37 chars omitted)
child 0, intervention: string
child 1, outcome: string
child 2, relationship: string
child 3, evidence_level: string
child 4, risk: string
child 5, source_ids: list<item: string>
child 0, item: string
child 4, plan: list<item: struct<intervention: string, framework: string, evidence_level: string, risk: string, pos (... 15 chars omitted)
child 0, item: struct<intervention: string, framework: string, evidence_level: string, risk: string, position: stri (... 3 chars omitted)
child 0, intervention: string
child 1, framework: string
child 2, evidence_level: string
child 3, risk: string
child 4, position: string
child 5, coach_reply: string
usage: struct<prompt_tokens: int64, completion_tokens: int64, cost_usd: double>
child 0, prompt_tokens: int64
child 1, completion_tokens: int64
child 2, cost_usd: double
generation_id: 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
to
{'messages': List({'role': Value('string'), 'content': 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 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
scenario_id: string
family_id: string
user_message: string
context: struct<bleeding_reported: bool, vomiting_persistent: bool, fever_reported: bool, weight_loss_lb: int (... 160 chars omitted)
child 0, bleeding_reported: bool
child 1, vomiting_persistent: bool
child 2, fever_reported: bool
child 3, weight_loss_lb: int64
child 4, weight_loss_intentional: bool
child 5, inability_to_pass_gas: bool
child 6, severe_progressive_pain: bool
child 7, melena: bool
child 8, abdominal_pain: bool
child 9, iron_deficiency_anemia: bool
gold: struct<care_level: string, red_flags: list<item: string>, tcm_pattern: string, arbitration: list<ite (... 177 chars omitted)
child 0, care_level: string
child 1, red_flags: list<item: string>
child 0, item: string
child 2, tcm_pattern: string
child 3, arbitration: list<item: struct<intervention: string, outcome: string, relationship: string, evidence_level: strin (... 3 chars omitted)
child 0, item: struct<intervention: string, outcome: string, relationship: string, evidence_level: string>
child 0, intervention: string
child 1, outcome: string
child 2, relationship: string
child 3, evidence_level: string
child 4, forbidden: list<item: struct<intervention: string, as: string, reason: string>>
child 0, item: struct<intervention: string, as: string, reason: string>
child 0, intervention: string
child 1, as: string
child 2, reason: string
cl
...
ities: list<item: string>
child 0, item: string
child 3, arbitration: list<item: struct<intervention: string, outcome: string, relationship: string, evidence_level: strin (... 49 chars omitted)
child 0, item: struct<intervention: string, outcome: string, relationship: string, evidence_level: string, risk: st (... 37 chars omitted)
child 0, intervention: string
child 1, outcome: string
child 2, relationship: string
child 3, evidence_level: string
child 4, risk: string
child 5, source_ids: list<item: string>
child 0, item: string
child 4, plan: list<item: struct<intervention: string, framework: string, evidence_level: string, risk: string, pos (... 15 chars omitted)
child 0, item: struct<intervention: string, framework: string, evidence_level: string, risk: string, position: stri (... 3 chars omitted)
child 0, intervention: string
child 1, framework: string
child 2, evidence_level: string
child 3, risk: string
child 4, position: string
child 5, coach_reply: string
usage: struct<prompt_tokens: int64, completion_tokens: int64, cost_usd: double>
child 0, prompt_tokens: int64
child 1, completion_tokens: int64
child 2, cost_usd: double
generation_id: 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
to
{'messages': List({'role': Value('string'), 'content': 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 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 datasetNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
messages list |
|---|
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
[
{
"role": "system",
"content": "You are an integrative digestive wellness coach. You interpret gut-health concerns through BOTH modern digestive medicine and Traditional Chinese Medicine, and you reconcile them with an explicit evidence-and-risk policy.\n\nReply with a SINGLE JSON object and nothing else:\n... |
End of preview.
Digestive Wellness Coach — training dataset (v4; v2 = sweep baseline)
The dataset is the deliverable (RUBRIC-02): 862 instruction rows teaching one falsifiable behavior — a digestive-wellness coach that arbitrates between TCM and modern evidence per intervention, emits a strict JSON policy (care level, per- intervention arbitration with evidence levels, safety gate), and never overstates evidence. Built from a 54-claim source-verified registry (ACG/AGA/Cochrane/NCCIH + user-supplied TCM primary texts with page provenance), never invented from thin air; teacher wrote prose only, labels are constructed (R24).
Files
train_v4.jsonl— 1205 rows, current dataset (v2 + round-1 pair topup + round-2 buried-alarm topup; see docs/t10d-diagnosis.md)train_v2.jsonl— 862 rows, the data-efficiency-sweep baselinecolab/train.jsonl,colab/valid.jsonl— chat-format splits (781/81, family-disjoint per R29)eval_scenarios.jsonl— held-out eval set: 40 scenarios, 16 families, family-disjoint from training; 6 adversarial families define robustnessclaims.json— the claims registry (54 claims, primary-source provenance)patterns.json— TCM pattern KB (8 weighted sign-maps, page provenance)source-registry.json— every source id cited by the spec
Provenance & code
Generation + quality-gate code (eval/build_dataset.py, eval/topup_dataset.py),
the 63-requirement PRD, and the eval harness live in the project git repo at commit
75863851837a7cc853b7052704e8ef7b6f7f8690 (accompanies the submission). Generation cost ≈ $0.0064/example.
Models trained on this dataset (Unsloth QLoRA on Qwen/Qwen3-1.7B, merged fp16)
| N | repo | weights commit | eval loss | spec adherence |
|---|---|---|---|---|
| 1102 (v4, T10d round 2) | kelsbeans/qwen3-1.7b-digestive-coach-n1102 | ffc4be4619c247b4ca5a2cb7d84ea5cd00faed8c |
0.7831 | 25.0% |
| 781 | kelsbeans/qwen3-1.7b-digestive-coach-n781 | 0b008de7d11cfab1b4c7a135dbb91b51cb8dc6b9 |
0.8283 | 20.0% |
| 390 | kelsbeans/qwen3-1.7b-digestive-coach-n390 | 6635e733bf488f8d693944de665cddcae2548eae |
0.9861 | 10.0% |
| 195 | kelsbeans/qwen3-1.7b-digestive-coach-n195 | c10602623262a8d87b6e468257c37b0647256d5c |
1.1959 | 10.0% |
| 97 | kelsbeans/qwen3-1.7b-digestive-coach-n97 | 74163732c0527d55ae8a754204abb606c8f98ade |
1.4385 | 7.5% |
Base model scores 0.0% spec adherence / 2.5% schema validity on the same eval;
the best prompted frontier cell reached 15% (prompt-ceiling ablation). Full curve +
minimum-viable-N analysis: docs/data-efficiency.md in the project repo.
- Downloads last month
- 60