Execution attempt · A17 · planned
Builds the GSM8K-PL, LLMzSzŁ STEM and PES item sets and the few-shot exemplars at pinned dataset revisions, after the gold-answer parity check.
Pinned source and configuration
https://github.com/stw2/tokenizer-science-tax @ c10ec1755e97a51ccb015ed8d566cfc1e50b4710
Reference checked 2026-09-14 13:10 UTC. Later commits, branches or plan changes do not retarget this attempt.
- Command
- python scripts/01_build_items.py
- Working directory
- experiments/E03-reasoning-language
- Configuration paths
- None
- Parameters
- seed 20260703; 250 GSM8K-PL, 300 LLMzSzŁ STEM and 250 PES items; 5 GSM8K-PL and 3 multiple-choice exemplars per benchmark
- Environment
- 3.12.13; CPU; Apple silicon, 128 GB unified memory
- Output directory
- Not recorded
Inputs
Materials the registrant named when registering this attempt, by content identity; obtainability is derived from their location reports. Nothing is fetched or verified.
- gsm8kx Polish test split · item source · M87 gsm8kx Polish GSM8K test split · dataset · Download
- gsm8kx Polish few-shot pool · item source · M88 gsm8kx Polish GSM8K few-shot pool · dataset · Download
- GSM8K main configuration · item source · M89 GSM8K main configuration · dataset · Download
- LLMzSzŁ test split · item source · M90 LLMzSzŁ test split · dataset · Download
- PES 2018-2022 questions · item source · M16 PES 2018-2022 questions · dataset · Download
- GSM8K bridge specification · item selection · M78 gsm8k_bridge_spec.json · raw output · Download
- authored exemplar traces · few shot traces · M91 Authored exemplar reasoning traces · dataset · Ask the reporter
Delivered events
Registered
#1Builds the GSM8K-PL, LLMzSzŁ STEM and PES item sets and the few-shot exemplars at pinned dataset revisions, after the gold-answer parity check.
Started
#2Started.
Succeeded
#3Completed with exit code 0.
Exit code 0.
- items_gsm8k_pl.jsonl · https://github.com/stw2/tokenizer-science-tax/blob/c10ec1755e97a51ccb015ed8d566cfc1e50b4710/experiments/E03-reasoning-language/data/items_gsm8k_pl.jsonl · public · sha256 49045ffff982… · 100500 bytes · M92
- items_llmzszl_stem.jsonl · Not redistributed: ask the Room owner, or rebuild with scripts/01_build_items.py at the pinned revisions · restricted · sha256 31bfda61690a… · 134994 bytes · M93
- items_pes.jsonl · Not redistributed: ask the Room owner, or rebuild with scripts/01_build_items.py at the pinned revisions · restricted · sha256 2b21c374fc95… · 170276 bytes · M94
- exemplars.json · Not redistributed: ask the Room owner, or rebuild with scripts/01_build_items.py at the pinned revisions · restricted · sha256 b69a61353968… · 12006 bytes · M95
- manifest.json · https://github.com/stw2/tokenizer-science-tax/blob/c10ec1755e97a51ccb015ed8d566cfc1e50b4710/experiments/E03-reasoning-language/data/manifest.json · public · sha256 8e8c876b13d1… · 10329 bytes · M96
Report an event
The registrant’s capture tool normally delivers start and outcome events. Reporting here is the same author report with the browser as the reported time; it does not observe the process.
This attempt has a delivered outcome. A new execution is a new attempt.