Execution attempt · A118 · planned
Derives the forward and reverse discordant translation pairs from the committed base-model scores, gated to the expected counts, and freezes the scan subsample, derangement, breakage guard and entity spans.
Pinned source and configuration
https://github.com/stw2/tokenizer-science-tax @ 7253735a8d117a39ade0ca39443d0726d77919fc
Reference checked 2026-09-14 13:17 UTC. Later commits, branches or plan changes do not retarget this attempt.
- Command
- python scripts/01_derive_sets.py
- Working directory
- experiments/E12-activation-patching
- Configuration paths
- None
- Parameters
- seed 20260703; scan subsample 50 per benchmark; breakage guard 50 per benchmark; endpoint correct_lennorm
- Environment
- 3.12.13; numpy on 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.
- E11 belebele_en.jsonl · evaluation data · M485 belebele_en.jsonl · raw output · Download
- E11 belebele_pl.jsonl · evaluation data · M486 belebele_pl.jsonl · raw output · Download
- E11 mmlu_en.jsonl · evaluation data · M487 mmlu_en.jsonl · raw output · Download
- E11 mmlu_pl.jsonl · evaluation data · M488 mmlu_pl.jsonl · raw output · Download
- E11 exclusions.json · evaluation data · M515 exclusions.json (13 excluded item pairs) · configuration · Download
- E11 scores_base.jsonl · reference scores · M496 scores_base.jsonl · raw output · Download
Delivered events
Registered
#1Derives the forward and reverse discordant translation pairs from the committed base-model scores, gated to the expected counts, and freezes the scan subsample, derangement, breakage guard and entity spans.
Started
#2Started.
Succeeded
#3Completed with exit code 0.
Exit code 0.
- sets.json · https://github.com/stw2/tokenizer-science-tax/blob/633b3ba1ce916e432ae5ba4e0a706ad58b53b2e9/experiments/E12-activation-patching/results/sets.json · public · sha256 a51c3c53c0ad… · 68374 bytes · M518
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.