sora 370953729b Fix perf stats (wrong import path), per-repeat checkpoints, README
- perf_stats aggregator lives in eval/, not model/: the import failed
  silently and EVERY perf column was empty (not just ttft). Now warns
  on stderr instead of swallowing.
- repeats > 1 get their own checkpoint key (:rep2, :rep3, ...): repeat 2
  previously restored repeat 1's predictions and finished instantly with
  identical scores. rep1 keeps the legacy key (existing checkpoints still
  resume).
- repeats summary: report the MEAN score and aggregate time/tokens over
  ALL runs (was: last run only).
- README: six-benchmark command as the primary example.

Co-Authored-By: Claude <noreply@anthropic.com>
2026-09-11 13:38:04 +00:00

41 lines
1.4 KiB
Python

"""BIG-Bench Hard (standard mirror: lukaemon/bbh; original: github.com/suzgunmirac/BIG-Bench-Hard).
Paper-faithful 3-shot CoT: the official hand-written exemplars (vendored in
_bbh_cot_prompts.py, MIT) are injected per-subtask via few_shot hook.
"""
from ..sample import Sample
from ..registry import register_dataset
from ..spec import DatasetSpec
def bbh_few_shot(split: str, subset: str, n: int):
"""Return the official 3-shot CoT prompt text for this subtask."""
if n <= 0:
return None
from ._bbh_cot_prompts import COT_PROMPTS
text = COT_PROMPTS.get(subset)
return text.strip() + '\n\n' if text else None
@register_dataset(
DatasetSpec(
name='bbh',
source='lukaemon/bbh', # https://huggingface.co/datasets/lukaemon/bbh
subset='boolean_expressions', # 27 subtasks; override with --subset <subtask>
split='test',
task_type='qa',
tags=['reasoning'],
description='BIG-Bench Hard, 27 subtasks (each subset caches under bbh/<hash>).',
prompt_style='bbh_es', # es test-question wrapper: Q:/A: think-step-by-step
few_shot_split='official_cot', # -> bbh_few_shot hook (official CoT)
few_shot_num=3, # paper/es default: 3-shot
)
)
def bbh():
def to_sample(record: dict) -> Sample:
return Sample(input=record['input'], target=str(record['target']).strip())
return to_sample