Execution attempt

Sign in with GitHub
← Experiment E8 · Does Bielik-PL-11B-v3.0-Instruct still pass through an English-like latent space when it reasons in Polish on Polish mathematics and STEM problems, and does the strength of that pivot predict whether its answer is correct?

Execution attempt · A77 · planned

Builds the English, Polish and other vocabulary partitions of both tokenizers from five corpora and checks golden pieces.

Succeeded · Registered by @stw2 via agent. Reported and received times are kept apart; no computation or result is verified.

Pinned source and configuration

https://github.com/stw2/tokenizer-science-tax @ 6a20c5c41c509137b0de80a80ec4fd3d3243ba56

Reference checked 2026-09-14 13:12 UTC. Later commits, branches or plan changes do not retarget this attempt.

Command
python scripts/01_build_lexicons.py
Working directory
experiments/E08-latent-pivot
Configuration paths
None
Parameters
min_count 5; rate_ratio 10; Polish-diacritic override; 150 traces per condition for the label mix; seed 20260704 for the audit samples
Environment
3.12.13; tokenizers through mlx_lm.utils.load_tokenizer; 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.

Delivered events

  1. Registered

    #1

    Builds the English, Polish and other vocabulary partitions of both tokenizers from five corpora and checks golden pieces.

    reported · received · @stw2 via agent · posted to the Thread

  2. Started

    #2

    Started.

    reported · received · @stw2 via agent · attempt only

  3. Succeeded

    #3

    Completed with exit code 0.

    Exit code 0.

    • lexicon_bielik-pl-11b-v3.0-instruct.json · https://github.com/stw2/tokenizer-science-tax/blob/6a20c5c41c509137b0de80a80ec4fd3d3243ba56/experiments/E08-latent-pivot/data/lexicon_bielik-pl-11b-v3.0-instruct.json · public · sha256 81f559e815cb… · 96712 bytes · M398
    • lexicon_bielik-pl-11b-v3.0-instruct_audit.txt · https://github.com/stw2/tokenizer-science-tax/blob/6a20c5c41c509137b0de80a80ec4fd3d3243ba56/experiments/E08-latent-pivot/data/lexicon_bielik-pl-11b-v3.0-instruct_audit.txt · public · sha256 a02ee690501c… · 1841 bytes · M399
    • lexicon_bielik-11b-v3.0-instruct.json · https://github.com/stw2/tokenizer-science-tax/blob/6a20c5c41c509137b0de80a80ec4fd3d3243ba56/experiments/E08-latent-pivot/data/lexicon_bielik-11b-v3.0-instruct.json · public · sha256 e902a73015b2… · 97095 bytes · M400
    • lexicon_bielik-11b-v3.0-instruct_audit.txt · https://github.com/stw2/tokenizer-science-tax/blob/6a20c5c41c509137b0de80a80ec4fd3d3243ba56/experiments/E08-latent-pivot/data/lexicon_bielik-11b-v3.0-instruct_audit.txt · public · sha256 4e3dd9dc0862… · 1685 bytes · M401
    • lexicons_report.json · https://github.com/stw2/tokenizer-science-tax/blob/6a20c5c41c509137b0de80a80ec4fd3d3243ba56/experiments/E08-latent-pivot/data/lexicons_report.json · public · sha256 a262abef4ba5… · 1167 bytes · M402

    reported · received · @stw2 via agent · posted to the Thread

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.