Datasets:

andyzhang232
/

ajev-lora7-decision-index

Tags:

License:

Dataset card Files Files and versions

xet

Community

Dataset Viewer

Duplicate

Split (1)

train

The dataset viewer is not available for this split.

Cannot load the dataset split (in streaming mode) to extract the first rows.

Error code:   StreamingRowsError
Exception:    CastError
Message:      Couldn't cast
edition: string
panel_id: string
index: double
raw_index: double
scores: struct<balanced_skill: double, balanced_raw: double, breadth_skill: double>
  child 0, balanced_skill: double
  child 1, balanced_raw: double
  child 2, breadth_skill: double
areas: list<item: struct<id: string, label: string, raw: double, skill: double, coverage: double, n: int64, (... 32 chars omitted)
  child 0, item: struct<id: string, label: string, raw: double, skill: double, coverage: double, n: int64, benchmarks (... 20 chars omitted)
      child 0, id: string
      child 1, label: string
      child 2, raw: double
      child 3, skill: double
      child 4, coverage: double
      child 5, n: int64
      child 6, benchmarks: list<item: int64>
          child 0, item: int64
benchmarks: struct<1: struct<raw: double, skill: double, coverage: double, random: double, rule: string, in_inde (... 5013 chars omitted)
  child 0, 1: struct<raw: double, skill: double, coverage: double, random: double, rule: string, in_index: bool, t (... 24 chars omitted)
      child 0, raw: double
      child 1, skill: double
      child 2, coverage: double
      child 3, random: double
      child 4, rule: string
      child 5, in_index: bool
      child 6, tracks: list<item: null>
          child 0, item: null
  child 1, 2: struct<raw: double, skill: double, coverage: double, random: double, rule: string, in_index: bool, t (... 24 chars omitted)
      child 0, raw: double
      child 1, skill: double
      child 2, covera
...
le
      child 3, random: double
      child 4, rule: string
      child 5, in_index: bool
  child 40, 24: struct<raw: double, skill: double, coverage: double, random: double, rule: string, in_index: bool>
      child 0, raw: double
      child 1, skill: double
      child 2, coverage: double
      child 3, random: double
      child 4, rule: string
      child 5, in_index: bool
  child 41, 26: struct<raw: double, skill: double, coverage: double, random: double, rule: string, in_index: bool>
      child 0, raw: double
      child 1, skill: double
      child 2, coverage: double
      child 3, random: double
      child 4, rule: string
      child 5, in_index: bool
  child 42, 27: struct<raw: double, skill: double, coverage: double, random: double, rule: string, in_index: bool>
      child 0, raw: double
      child 1, skill: double
      child 2, coverage: double
      child 3, random: double
      child 4, rule: string
      child 5, in_index: bool
  child 43, 34: struct<raw: double, skill: double, coverage: double, random: double, rule: string, in_index: bool>
      child 0, raw: double
      child 1, skill: double
      child 2, coverage: double
      child 3, random: double
      child 4, rule: string
      child 5, in_index: bool
coverage: double
note: string
successful_request_latency_ms: struct<median: double, p95: double, mean: double>
  child 0, median: double
  child 1, p95: double
  child 2, mean: double
engine: string
counts: struct<ok: int64>
  child 0, ok: int64
to
{'engine': Value('string'), 'counts': {'ok': Value('int64')}, 'successful_request_latency_ms': {'median': Value('float64'), 'p95': Value('float64'), 'mean': Value('float64')}, 'benchmarks': List({'catalog_id': Value('int64'), 'dataset': Value('string'), 'requests': Value('int64'), 'answered': Value('int64'), 'unsupported': Value('int64'), 'errors': Value('int64'), 'abstained': Value('int64'), 'pending': Value('int64'), 'scored_requests': Value('int64'), 'metric': Value('string'), 'score': Value('float64'), 'reference_same_cases': Value('null'), 'median_ms': Value('float64'), 'detail': {'field_accuracy': Value('float64'), 'case_exact_accuracy': Value('float64'), 'custom_metrics': {'ndcg_at_10': Value('float64'), 'mrr': Value('float64'), 'recall_at_10': Value('float64'), 'candidate_recall': Value('float64'), 'scorable_candidate_recall': Value('float64'), 'candidates_scored': Value('float64'), 'candidates_retrieved': Value('int64'), 'bm25_ndcg_at_10': Value('float64'), 'quality_quality': Value('float64'), 'quality_cost_usd': Value('float64'), 'quality_utility': Value('float64'), 'quality_oracle_optimal': Value('float64'), 'quality_utility_regret': Value('float64'), 'cost_aware_quality': Value('float64'), 'cost_aware_cost_usd': Value('float64'), 'cost_aware_utility': Value('float64'), 'cost_aware_oracle_optimal': Value('float64'), 'cost_aware_utility_regret': Value('float64'), 'value_regret': Value('float64'), 'brier': Value('float64'), 'log_loss': Value('float64'), 'accuracy': V
...
lue('int64'), 'mean': Value('float64')}}, 'cluster_macro_accuracy': Value('float64'), 'positive_f1_by_field': {'sarcastic': Value('float64'), 'sarcasm': Value('float64'), 'irony': Value('float64'), 'satire': Value('float64'), 'understatement': Value('float64'), 'overstatement': Value('float64'), 'rhetorical_question': Value('float64')}, 'category_macro_f1': Value('float64'), 'scored_fields': Value('int64'), 'chance_on_rows': Value('float64')}, 'tracks': {'RouterBench-0shot': {'metric': Value('string'), 'score': Value('float64'), 'scored_requests': Value('int64')}, 'RouterBench-5shot': {'metric': Value('string'), 'score': Value('float64'), 'scored_requests': Value('int64')}, 'GSM8K-10choice': {'metric': Value('string'), 'score': Value('float64'), 'scored_requests': Value('int64')}, 'GSM8K-4choice': {'metric': Value('string'), 'score': Value('float64'), 'scored_requests': Value('int64')}, 'iSarcasmEval-A-Ar': {'metric': Value('string'), 'score': Value('null'), 'scored_requests': Value('int64')}, 'iSarcasmEval-A-En': {'metric': Value('string'), 'score': Value('null'), 'scored_requests': Value('int64')}, 'iSarcasmEval-B-En': {'metric': Value('string'), 'score': Value('null'), 'scored_requests': Value('int64')}, 'iSarcasmEval-C-Ar': {'metric': Value('string'), 'score': Value('null'), 'scored_requests': Value('int64')}, 'iSarcasmEval-C-En': {'metric': Value('string'), 'score': Value('null'), 'scored_requests': Value('int64')}}}), 'note': Value('string'), 'edition': Value('string')}
because column names don't match
Traceback:    Traceback (most recent call last):
                File "/src/services/worker/src/worker/utils.py", line 147, in get_rows_or_raise
                  return get_rows(
                      dataset=dataset,
                  ...<4 lines>...
                      column_names=column_names,
                  )
                File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
                  return func(*args, **kwargs)
                File "/src/services/worker/src/worker/utils.py", line 127, in get_rows
                  rows_plus_one = list(itertools.islice(safe_iter(ds, dataset=dataset), rows_max_number + 1))
                File "/src/services/worker/src/worker/utils.py", line 483, in safe_iter
                  yield from ds.decode(False) if ds.features else ds
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2951, in __iter__
                  for key, example in ex_iterable:
                                      ^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2461, in __iter__
                  for key, pa_table in self._iter_arrow():
                                       ~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2486, in _iter_arrow
                  for key, pa_table in self.ex_iterable._iter_arrow():
                                       ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 547, in _iter_arrow
                  for key, pa_table in iterator:
                                       ^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 430, in _iter_arrow
                  for key, pa_table in self.generate_tables_fn(**gen_kwags):
                                       ~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
                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
              edition: string
              panel_id: string
              index: double
              raw_index: double
              scores: struct<balanced_skill: double, balanced_raw: double, breadth_skill: double>
                child 0, balanced_skill: double
                child 1, balanced_raw: double
                child 2, breadth_skill: double
              areas: list<item: struct<id: string, label: string, raw: double, skill: double, coverage: double, n: int64, (... 32 chars omitted)
                child 0, item: struct<id: string, label: string, raw: double, skill: double, coverage: double, n: int64, benchmarks (... 20 chars omitted)
                    child 0, id: string
                    child 1, label: string
                    child 2, raw: double
                    child 3, skill: double
                    child 4, coverage: double
                    child 5, n: int64
                    child 6, benchmarks: list<item: int64>
                        child 0, item: int64
              benchmarks: struct<1: struct<raw: double, skill: double, coverage: double, random: double, rule: string, in_inde (... 5013 chars omitted)
                child 0, 1: struct<raw: double, skill: double, coverage: double, random: double, rule: string, in_index: bool, t (... 24 chars omitted)
                    child 0, raw: double
                    child 1, skill: double
                    child 2, coverage: double
                    child 3, random: double
                    child 4, rule: string
                    child 5, in_index: bool
                    child 6, tracks: list<item: null>
                        child 0, item: null
                child 1, 2: struct<raw: double, skill: double, coverage: double, random: double, rule: string, in_index: bool, t (... 24 chars omitted)
                    child 0, raw: double
                    child 1, skill: double
                    child 2, covera
              ...
              le
                    child 3, random: double
                    child 4, rule: string
                    child 5, in_index: bool
                child 40, 24: struct<raw: double, skill: double, coverage: double, random: double, rule: string, in_index: bool>
                    child 0, raw: double
                    child 1, skill: double
                    child 2, coverage: double
                    child 3, random: double
                    child 4, rule: string
                    child 5, in_index: bool
                child 41, 26: struct<raw: double, skill: double, coverage: double, random: double, rule: string, in_index: bool>
                    child 0, raw: double
                    child 1, skill: double
                    child 2, coverage: double
                    child 3, random: double
                    child 4, rule: string
                    child 5, in_index: bool
                child 42, 27: struct<raw: double, skill: double, coverage: double, random: double, rule: string, in_index: bool>
                    child 0, raw: double
                    child 1, skill: double
                    child 2, coverage: double
                    child 3, random: double
                    child 4, rule: string
                    child 5, in_index: bool
                child 43, 34: struct<raw: double, skill: double, coverage: double, random: double, rule: string, in_index: bool>
                    child 0, raw: double
                    child 1, skill: double
                    child 2, coverage: double
                    child 3, random: double
                    child 4, rule: string
                    child 5, in_index: bool
              coverage: double
              note: string
              successful_request_latency_ms: struct<median: double, p95: double, mean: double>
                child 0, median: double
                child 1, p95: double
                child 2, mean: double
              engine: string
              counts: struct<ok: int64>
                child 0, ok: int64
              to
              {'engine': Value('string'), 'counts': {'ok': Value('int64')}, 'successful_request_latency_ms': {'median': Value('float64'), 'p95': Value('float64'), 'mean': Value('float64')}, 'benchmarks': List({'catalog_id': Value('int64'), 'dataset': Value('string'), 'requests': Value('int64'), 'answered': Value('int64'), 'unsupported': Value('int64'), 'errors': Value('int64'), 'abstained': Value('int64'), 'pending': Value('int64'), 'scored_requests': Value('int64'), 'metric': Value('string'), 'score': Value('float64'), 'reference_same_cases': Value('null'), 'median_ms': Value('float64'), 'detail': {'field_accuracy': Value('float64'), 'case_exact_accuracy': Value('float64'), 'custom_metrics': {'ndcg_at_10': Value('float64'), 'mrr': Value('float64'), 'recall_at_10': Value('float64'), 'candidate_recall': Value('float64'), 'scorable_candidate_recall': Value('float64'), 'candidates_scored': Value('float64'), 'candidates_retrieved': Value('int64'), 'bm25_ndcg_at_10': Value('float64'), 'quality_quality': Value('float64'), 'quality_cost_usd': Value('float64'), 'quality_utility': Value('float64'), 'quality_oracle_optimal': Value('float64'), 'quality_utility_regret': Value('float64'), 'cost_aware_quality': Value('float64'), 'cost_aware_cost_usd': Value('float64'), 'cost_aware_utility': Value('float64'), 'cost_aware_oracle_optimal': Value('float64'), 'cost_aware_utility_regret': Value('float64'), 'value_regret': Value('float64'), 'brier': Value('float64'), 'log_loss': Value('float64'), 'accuracy': V
              ...
              lue('int64'), 'mean': Value('float64')}}, 'cluster_macro_accuracy': Value('float64'), 'positive_f1_by_field': {'sarcastic': Value('float64'), 'sarcasm': Value('float64'), 'irony': Value('float64'), 'satire': Value('float64'), 'understatement': Value('float64'), 'overstatement': Value('float64'), 'rhetorical_question': Value('float64')}, 'category_macro_f1': Value('float64'), 'scored_fields': Value('int64'), 'chance_on_rows': Value('float64')}, 'tracks': {'RouterBench-0shot': {'metric': Value('string'), 'score': Value('float64'), 'scored_requests': Value('int64')}, 'RouterBench-5shot': {'metric': Value('string'), 'score': Value('float64'), 'scored_requests': Value('int64')}, 'GSM8K-10choice': {'metric': Value('string'), 'score': Value('float64'), 'scored_requests': Value('int64')}, 'GSM8K-4choice': {'metric': Value('string'), 'score': Value('float64'), 'scored_requests': Value('int64')}, 'iSarcasmEval-A-Ar': {'metric': Value('string'), 'score': Value('null'), 'scored_requests': Value('int64')}, 'iSarcasmEval-A-En': {'metric': Value('string'), 'score': Value('null'), 'scored_requests': Value('int64')}, 'iSarcasmEval-B-En': {'metric': Value('string'), 'score': Value('null'), 'scored_requests': Value('int64')}, 'iSarcasmEval-C-Ar': {'metric': Value('string'), 'score': Value('null'), 'scored_requests': Value('int64')}, 'iSarcasmEval-C-En': {'metric': Value('string'), 'score': Value('null'), 'scored_requests': Value('int64')}}}), 'note': Value('string'), 'edition': Value('string')}
              because column names don't match

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.

AJev lora7 · Decision Index 0.2.1 results

Self-run results of andyzhang232/ajev-gemma4-12b-lora7 (Gemma 4 12B + LoRA) on the Jev Decision Index, produced and scored with the official reproduction kit apolinario/decision-index (commit 87d4650, edition 0.2.1).


Decision Index (chance-corrected)	55.06
Raw index	65.15
Requests answered	150,317 / 150,317 scoreable (`complete: true`, all `ok`; no unsupported, error or abstained rows)
Areas (skill × 100)	Knowledge & Reasoning 38.6 · Language Understanding 63.4 · Retrieval & Classification 63.0 · Tools & Automation 67.5 · Arts & Human Taste 37.3

Files

File	Contents
`scores.json`	the kit's combined scores file (the one a submission points at)
`index.json`	per-benchmark raw / skill / coverage / chance and the five areas
`benchmark-summary.json`	native metric per benchmark with answered / unsupported / error counts
`results.compact.jsonl.gz`	one line per request in the kit's result format, with `payload` and `raw_output` removed (the suite's source texts are not redistributed; every row keeps its `payload_sha256`)
`runners/`	each official runner's `environment.json` and final `status.json`
`harness/`	the sharding, merge and scoring scripts and a step-by-step `RUN.md`

How it was run

Engine. ajev-infer @ bbfbf7f (0.6.0; the inference modules are unchanged from 0.5.0, which ran the lora5 submission): the same prompt and label readout as everyday use. Every request is sent whole, with no truncation, no option pruning and no benchmark-specific prompt.
- All 150,317 requests, HLE included, went through the kit's http engine against python -m ajev.serve_vllm (vLLM 0.30.0, bf16, LoRA loaded directly, max_model_len 131,072) with 8 official runners per server on Colab RTX PRO 6000 (96 GB) GPUs.
- The first machine was reclaimed after 36,155 requests; the rest were split in two halves and run on two machines at once. Results were merged by run_id, keeping the ok row when a request had been re-run (see harness/RUN.md).
- No separate vLLM-vs-transformers check was run for this adapter. The serving code and vLLM version are the same as for lora5, whose check on 600 decisions gave identical prompt tokens (600/600), 98.2% argmax agreement and mean |Δp| 0.0115.
Suite. The same rebuilt suite as the lora5 run (see andyzhang232/ajev-lora5-decision-index): the seven added benchmarks are byte-identical to the lab's file; the base rows were rebuilt without HLE and cut with the kit's release_v2.cut; HLE was rebuilt separately and its 513 group ids match hub/manifest.json.
Latency. The latency_ms in scores.json (median 213 ms) was measured with 8 requests in flight per server, so it includes queueing and is not a single-request figure. Single requests, one at a time on one RTX PRO 6000, take about 40 ms on the vLLM server for this 12B architecture (measured on lora5, same base and LoRA shape).

Reproduce

pip install "ajev-infer[leaderboard] @ git+https://github.com/cmzy/ajev-infer"
python -m decision_index run --engine ajev.jdi_engine:AJevEngine \
    --option model=google/gemma-4-12B-it --option adapter=andyzhang232/ajev-gemma4-12b-lora7 --out runs/ajev-lora7
python -m decision_index score --results runs/ajev-lora7/results.jsonl --edition 0.2.1

Training used only train splits and was decontaminated against the suite's evaluation data (exact text and shared ≥ 60-character sentences), and every new data source was checked for answer shortcuts on its own dev split. Not affiliated with TypeSafe AI.

Downloads last month: 78

Total file size:

41.2 MB