Dataset Preview
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed because of a cast error
Error code: DatasetGenerationCastError
Exception: DatasetGenerationCastError
Message: An error occurred while generating the dataset
All the data files must have the same columns, but at some point there are 4 new columns ({'answered_by', 'gp_was_wrong', 'gp_flagged', 'escalated'}) and 14 missing columns ({'model', 'answer', 'call_error', 'seconds', 'cached', 'hedges', 'unsafe_claims', 'prompt_tokens', 'bonus_hits', 'refused', 'words', 'reading_grade', 'missed', 'completion_tokens'}).
This happened while the csv dataset builder was generating data using
hf://datasets/richardyoung/lab04-gp-vs-specialist/cascade.csv (at revision a40d0c6cc3b3b415b2b7e577860ecad682a9a33e), ['hf://datasets/richardyoung/lab04-gp-vs-specialist@a40d0c6cc3b3b415b2b7e577860ecad682a9a33e/answers.csv', 'hf://datasets/richardyoung/lab04-gp-vs-specialist@a40d0c6cc3b3b415b2b7e577860ecad682a9a33e/cascade.csv', 'hf://datasets/richardyoung/lab04-gp-vs-specialist@a40d0c6cc3b3b415b2b7e577860ecad682a9a33e/partA_temperature_sweep.csv', 'hf://datasets/richardyoung/lab04-gp-vs-specialist@a40d0c6cc3b3b415b2b7e577860ecad682a9a33e/per_question.csv', 'hf://datasets/richardyoung/lab04-gp-vs-specialist@a40d0c6cc3b3b415b2b7e577860ecad682a9a33e/scorecard.csv']
Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1848, in _prepare_split_single
writer.write_table(table)
~~~~~~~~~~~~~~~~~~^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/arrow_writer.py", line 765, in write_table
self._write_table(pa_table, writer_batch_size=writer_batch_size)
~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/arrow_writer.py", line 773, in _write_table
pa_table = table_cast(pa_table, self._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
question: string
domain: string
escalated: bool
answered_by: string
coverage: double
full_credit: int64
cost_usd: double
gp_flagged: bool
gp_was_wrong: bool
-- schema metadata --
pandas: '{"index_columns": [{"kind": "range", "name": null, "start": 0, "' + 1327
to
{'question': Value('string'), 'domain': Value('string'), 'model': Value('string'), 'answer': Value('string'), 'seconds': Value('float64'), 'prompt_tokens': Value('int64'), 'completion_tokens': Value('int64'), 'cost_usd': Value('float64'), 'cached': Value('bool'), 'call_error': Value('string'), 'coverage': Value('float64'), 'full_credit': Value('int64'), 'missed': Value('string'), 'unsafe_claims': Value('float64'), 'bonus_hits': Value('string'), 'hedges': Value('bool'), 'refused': Value('bool'), 'words': Value('int64'), 'reading_grade': Value('float64')}
because column names don't match
During handling of the above exception, another exception occurred:
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 1850, in _prepare_split_single
raise DatasetGenerationCastError.from_cast_error(
...<4 lines>...
)
datasets.exceptions.DatasetGenerationCastError: An error occurred while generating the dataset
All the data files must have the same columns, but at some point there are 4 new columns ({'answered_by', 'gp_was_wrong', 'gp_flagged', 'escalated'}) and 14 missing columns ({'model', 'answer', 'call_error', 'seconds', 'cached', 'hedges', 'unsafe_claims', 'prompt_tokens', 'bonus_hits', 'refused', 'words', 'reading_grade', 'missed', 'completion_tokens'}).
This happened while the csv dataset builder was generating data using
hf://datasets/richardyoung/lab04-gp-vs-specialist/cascade.csv (at revision a40d0c6cc3b3b415b2b7e577860ecad682a9a33e), ['hf://datasets/richardyoung/lab04-gp-vs-specialist@a40d0c6cc3b3b415b2b7e577860ecad682a9a33e/answers.csv', 'hf://datasets/richardyoung/lab04-gp-vs-specialist@a40d0c6cc3b3b415b2b7e577860ecad682a9a33e/cascade.csv', 'hf://datasets/richardyoung/lab04-gp-vs-specialist@a40d0c6cc3b3b415b2b7e577860ecad682a9a33e/partA_temperature_sweep.csv', 'hf://datasets/richardyoung/lab04-gp-vs-specialist@a40d0c6cc3b3b415b2b7e577860ecad682a9a33e/per_question.csv', 'hf://datasets/richardyoung/lab04-gp-vs-specialist@a40d0c6cc3b3b415b2b7e577860ecad682a9a33e/scorecard.csv']
Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)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.
question string | domain string | model string | answer string | seconds float64 | prompt_tokens int64 | completion_tokens int64 | cost_usd float64 | cached bool | call_error string | coverage float64 | full_credit int64 | missed string | unsafe_claims null | bonus_hits string | hedges bool | refused bool | words int64 | reading_grade float64 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
onc_staging | Oncology staging | liquid/lfm-2.5-2.6b:free | null | 6.42 | 125 | 2,000 | 0 | false | empty answer: all 2000 completion tokens went to reasoning (2000). Raise MAX_TOKENS. | 0 | 0 | t2|n1|m0|iib | null | null | false | false | 0 | null |
cardio_gdm | Cardiology guideline | liquid/lfm-2.5-2.6b:free | The four drug classes that constitute guideline-directed medical therapy (GDMT) for HFrEF are:
1. **ACE inhibitors** (or ARBs/ARNIs if ACEi contraindicated)
2. **Beta-blockers** (e.g., carvedilol, metoprolol succinate, bisoprolol)
3. **Mineralocorticoid receptor antagonists** (e.g., spironolactone, eplerenone)
4. **SG... | 3.07 | 122 | 897 | 0 | false | null | 1 | 1 | null | null | null | true | false | 138 | 13.1 |
ddi_warfarin | Drug-drug interaction | liquid/lfm-2.5-2.6b:free | **Interaction Analysis**
**1. Fluconazole + Warfarin**
- Fluconazole is a moderate inhibitor of CYP2C9, the primary enzyme responsible for metabolizing S-warfarin (the more active enantiomer).
- Inhibition of CYP2C9 by fluconazole leads to increased warfarin plasma concentrations and elevated INR.
- Clinical consequen... | 8.66 | 120 | 1,673 | 0 | false | null | 1 | 1 | null | null | cyp2c9|cyp3a4 | true | false | 310 | 12.2 |
rare_bone | Rare-disease workup | liquid/lfm-2.5-2.6b:free | **Most Likely Diagnosis:** Autosomal dominant osteopetrosis (ADOPT)
**Genes Most Often Involved:**
- **TP53** — the most common cause (~50% of cases); accounts for the majority of autosomal dominant forms
- Other genes (less common): *LRRC1* (autosomal recessive), *SLC37A3* (autosomal dominant), *NCOA4* (autosomal dom... | 2.14 | 119 | 569 | 0 | false | null | 0.333 | 0 | osteogenesis imperfecta|col1a1 | null | autosomal dominant | true | false | 140 | 36.7 |
onto_map | Ontology mapping (ICD-10-CM to SNOMED CT) | liquid/lfm-2.5-2.6b:free | <|tool_call_start|>[google(query='ICD-10-CM E11.9 Type 2 diabetes mellitus without complications SNOMED CT mapping SCTID'), google(query='SNOMED CT ICD-10 crosswalk E11.9')]<|tool_call_end|> | 0.99 | 144 | 192 | 0 | false | null | 0.5 | 0 | granular | null | map | false | false | 28 | 6.2 |
onc_staging | Oncology staging | nvidia/nemotron-3-ultra-550b-a55b:free | **T category: T2** (tumor >2 cm but ≤5 cm; 2.5 cm invasive ductal carcinoma confined to breast)
**N category: N1** (metastases in 1–3 axillary lymph nodes; two positive sentinel nodes)
**M category: M0** (no distant metastases on imaging)
**AJCC Anatomic Stage Group: Stage IIB** (T2 N1 M0)
*Reference: AJCC Canc... | 41.26 | 123 | 605 | 0 | false | null | 1 | 1 | null | null | sentinel | false | false | 53 | 13.8 |
cardio_gdm | Cardiology guideline | nvidia/nemotron-3-ultra-550b-a55b:free | **Four GDMT pillars for HFrEF (ACC/AHA/HFSA 2022, ESC 2021):**
1. **ARNI (preferred) / ACEi / ARB** — renin–angiotensin inhibition
2. **Beta-blocker** (carvedilol, metoprolol succinate, bisoprolol)
3. **MRA** (spironolactone or eplerenone)
4. **SGLT2 inhibitor** (dapagliflozin or empagliflozin)
**What to change ... | 22.83 | 126 | 904 | 0 | false | null | 1 | 1 | null | null | null | true | false | 79 | 13.1 |
ddi_warfarin | Drug-drug interaction | nvidia/nemotron-3-ultra-550b-a55b:free | null | 0.46 | 0 | 0 | 0 | false | null | 0 | 0 | inr|bleed|clot|avoid | null | null | false | false | 0 | null |
rare_bone | Rare-disease workup | nvidia/nemotron-3-ultra-550b-a55b:free | **Diagnosis:** Osteogenesis imperfecta (OI), most likely **Type I** (mild, autosomal dominant) given the combination of recurrent fractures, blue-grey sclerae, dentinogenesis imperfecta, and multi-generational inheritance.
**Genes most often involved:**
- **COL1A1** (17q21.33) – ~90% of pathogenic variants in domina... | 6.8 | 120 | 400 | 0 | false | null | 1 | 1 | null | null | autosomal dominant | true | false | 120 | 15.2 |
onto_map | Ontology mapping (ICD-10-CM to SNOMED CT) | nvidia/nemotron-3-ultra-550b-a55b:free | **SNOMED CT concept:** Type 2 diabetes mellitus (disorder)
**SCTID:** 44054006
*Mapping source: NLM ICD-10-CM to SNOMED CT Map (2024 release).*
ICD-10-CM is a statistical classification that groups clinically distinct conditions into single codes for billing and epidemiology (e.g., “without complications” is a re... | 54.29 | 148 | 651 | 0 | false | null | 1 | 1 | null | null | 44054006|map | false | false | 90 | 15.3 |
onc_staging | Oncology staging | ~openai/gpt-astra-latest | Using **AJCC 8th-edition anatomic staging**, assuming the nodal metastases are **macrometastases (>2 mm)**:
- **T2:** Tumor >2 cm but ≤5 cm (here, 2.5 cm).
- **N1:** Metastases in 1–3 ipsilateral axillary nodes; specifically **pN1a(sn)** based on sentinel-node pathology.
- **cM0:** No clinical or imaging evidence of d... | 8.81 | 117 | 400 | 0.02117 | false | null | 1 | 1 | null | null | sentinel | false | false | 65 | 10.7 |
cardio_gdm | Cardiology guideline | ~openai/gpt-astra-latest | The **2022 AHA/ACC/HFSA heart-failure guideline** recommends four foundational drug classes for HFrEF:
1. **Renin–angiotensin system inhibition:** preferably an **ARNI** (sacubitril/valsartan); an ACE inhibitor or ARB if ARNI is not feasible.
2. **Evidence-based β-blocker:** carvedilol, metoprolol **succinate**, or bi... | 13.16 | 117 | 602 | 0.03127 | false | null | 1 | 1 | null | null | titrate|potassium | true | false | 162 | 13.2 |
ddi_warfarin | Drug-drug interaction | ~openai/gpt-astra-latest | **Avoid St John’s wort, and arrange a warfarin-management plan before or promptly after starting fluconazole.** A stable INR of 2.4 does not protect against these interactions.
| Combination | Mechanism and clinical consequence |
|---|---|
| **Fluconazole + warfarin** | Fluconazole inhibits **CYP2C9**, reducing cleara... | 16.04 | 111 | 660 | 0.03411 | false | null | 1 | 1 | null | null | cyp2c9|cyp3a4 | true | false | 252 | 14 |
rare_bone | Rare-disease workup | ~openai/gpt-astra-latest | - **Most likely diagnosis:** **Osteogenesis imperfecta (OI), with autosomal dominant inheritance**—supported by recurrent low-trauma fractures, blue-grey sclerae, dentinogenesis imperfecta, and affected individuals across three generations.
- **Genes most often involved:** **COL1A1** and **COL1A2**, encoding the α1 and... | 6.63 | 111 | 266 | 0.01441 | false | null | 1 | 1 | null | null | autosomal dominant | true | false | 93 | 18.1 |
onto_map | Ontology mapping (ICD-10-CM to SNOMED CT) | ~openai/gpt-astra-latest | **SNOMED CT:** *Type 2 diabetes mellitus without complication (disorder)* — **SCTID 313436004** — is the direct semantic match for **ICD-10-CM E11.9**; verify availability in your target SNOMED CT edition and release.
ICD-10-CM groups diagnoses for classification and reporting, whereas SNOMED CT represents clinical me... | 8.1 | 135 | 320 | 0.01735 | false | null | 1 | 1 | null | null | map | true | false | 82 | 14.3 |
onc_staging | Oncology staging | null | null | null | null | null | 0 | null | null | 0 | 0 | null | null | null | null | null | null | null |
cardio_gdm | Cardiology guideline | null | null | null | null | null | 0 | null | null | 1 | 1 | null | null | null | null | null | null | null |
ddi_warfarin | Drug-drug interaction | null | null | null | null | null | 0 | null | null | 1 | 1 | null | null | null | null | null | null | null |
rare_bone | Rare-disease workup | null | null | null | null | null | 0 | null | null | 0.333 | 0 | null | null | null | null | null | null | null |
onto_map | Ontology mapping (ICD-10-CM to SNOMED CT) | null | null | null | null | null | 0 | null | null | 0.5 | 0 | null | null | null | null | null | null | null |
null | null | null | null | 48.9 | 60 | null | null | null | null | null | null | null | null | null | null | null | null | null |
null | null | null | null | 24.1 | 60 | null | null | null | null | null | null | null | null | null | null | null | null | null |
null | null | null | null | 24.3 | 60 | null | null | null | null | null | null | null | null | null | null | null | null | null |
null | null | null | null | 25.7 | 60 | null | null | null | null | null | null | null | null | null | null | null | null | null |
null | null | null | null | 28 | 60 | null | null | null | null | null | null | null | null | null | null | null | null | null |
onc_staging | Oncology staging | liquid/lfm-2.5-2.6b:free | null | 6.42 | null | 2,000 | 0 | null | null | 0 | 0 | null | null | null | null | null | null | null |
cardio_gdm | Cardiology guideline | liquid/lfm-2.5-2.6b:free | null | 3.07 | null | 897 | 0 | null | null | 1 | 1 | null | null | null | null | null | null | null |
ddi_warfarin | Drug-drug interaction | liquid/lfm-2.5-2.6b:free | null | 8.66 | null | 1,673 | 0 | null | null | 1 | 1 | null | null | null | null | null | null | null |
rare_bone | Rare-disease workup | liquid/lfm-2.5-2.6b:free | null | 2.14 | null | 569 | 0 | null | null | 0.333 | 0 | null | null | null | null | null | null | null |
onto_map | Ontology mapping (ICD-10-CM to SNOMED CT) | liquid/lfm-2.5-2.6b:free | null | 0.99 | null | 192 | 0 | null | null | 0.5 | 0 | null | null | null | null | null | null | null |
onc_staging | Oncology staging | nvidia/nemotron-3-ultra-550b-a55b:free | null | 41.26 | null | 605 | 0 | null | null | 1 | 1 | null | null | null | null | null | null | null |
cardio_gdm | Cardiology guideline | nvidia/nemotron-3-ultra-550b-a55b:free | null | 22.83 | null | 904 | 0 | null | null | 1 | 1 | null | null | null | null | null | null | null |
ddi_warfarin | Drug-drug interaction | nvidia/nemotron-3-ultra-550b-a55b:free | null | 0.46 | null | 0 | 0 | null | null | 0 | 0 | null | null | null | null | null | null | null |
rare_bone | Rare-disease workup | nvidia/nemotron-3-ultra-550b-a55b:free | null | 6.8 | null | 400 | 0 | null | null | 1 | 1 | null | null | null | null | null | null | null |
onto_map | Ontology mapping (ICD-10-CM to SNOMED CT) | nvidia/nemotron-3-ultra-550b-a55b:free | null | 54.29 | null | 651 | 0 | null | null | 1 | 1 | null | null | null | null | null | null | null |
onc_staging | Oncology staging | ~openai/gpt-astra-latest | null | 8.81 | null | 400 | 0.02117 | null | null | 1 | 1 | null | null | null | null | null | null | null |
cardio_gdm | Cardiology guideline | ~openai/gpt-astra-latest | null | 13.16 | null | 602 | 0.03127 | null | null | 1 | 1 | null | null | null | null | null | null | null |
ddi_warfarin | Drug-drug interaction | ~openai/gpt-astra-latest | null | 16.04 | null | 660 | 0.03411 | null | null | 1 | 1 | null | null | null | null | null | null | null |
rare_bone | Rare-disease workup | ~openai/gpt-astra-latest | null | 6.63 | null | 266 | 0.01441 | null | null | 1 | 1 | null | null | null | null | null | null | null |
onto_map | Ontology mapping (ICD-10-CM to SNOMED CT) | ~openai/gpt-astra-latest | null | 8.1 | null | 320 | 0.01735 | null | null | 1 | 1 | null | null | null | null | null | null | null |
null | null | liquid/lfm-2.5-2.6b:free | null | null | null | null | null | null | null | 0.567 | 0.4 | null | null | null | null | null | null | 17.1 |
null | null | nvidia/nemotron-3-ultra-550b-a55b:free | null | null | null | null | null | null | null | 0.8 | 0.8 | null | null | null | null | null | null | 14.3 |
null | null | ~openai/gpt-astra-latest | null | null | null | null | null | null | null | 1 | 1 | null | null | null | null | null | null | 14.1 |
No dataset card yet
- Downloads last month
- 22