Execution attempt

Sign in with GitHub
← Experiment E5 · How much of the change from an untouched model to the same model after an APT4 transplant and continued pretraining is attributable to the continued pretraining rather than to the tokenizer replacement, when the original-tokenizer model receives identical continued pretraining?

Execution attempt · A49 · planned

Scores the digit probe on the transplant at time zero and arm B's five milestones.

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 @ b4c55e6e7d26917e239ccbe2f11886c4d4a733a2

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

Command
bash scripts/06_score_checkpoint.sh digits armB t0_fvt=ckpts/fvt_t0 ckpt_00100M=ckpts/armB/ckpt_00100M ckpt_00200M=ckpts/armB/ckpt_00200M ckpt_00300M=ckpts/armB/ckpt_00300M ckpt_00400M=ckpts/armB/ckpt_00400M ckpt_00500M=ckpts/armB/ckpt_00500M
Working directory
experiments/E05-control-arm
Configuration paths
None
Parameters
300 items in each of three formats, 4-shot plain-text prompts, greedy decoding with batch 24, completion cut at the first newline, graded by E2's dual-locale grader after its 31-case golden suite; BOS parity checked per model.
Environment
3.12.13; PyTorch MPS, bf16; 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

    Scores the digit probe on the transplant at time zero and arm B's five milestones.

    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.

    • probe_digits_armB.jsonl · https://github.com/stw2/tokenizer-science-tax/blob/9a57aa40f4e774136f51b6730c33266d59094a45/experiments/E05-control-arm/results/probe_digits_armB.jsonl · public · sha256 98c937522572… · 5518 bytes · M172
    • probe_digits_armB.jsonl as originally written · Archived private repository, experiments/E5/results/probe_digits_armB.jsonl at commit 5389273; the public copy replaces absolute model paths with paths relative to the experiment folder · unavailable · sha256 b0d59f45a42b… · 5830 bytes · M173
    • armB/t0_fvt.jsonl · https://github.com/stw2/tokenizer-science-tax/blob/9a57aa40f4e774136f51b6730c33266d59094a45/experiments/E05-control-arm/results/probe_digits_raw/armB/t0_fvt.jsonl · public · sha256 4d5bc26d2c62… · 263504 bytes · M174
    • armB/t0_fvt.jsonl as originally written · Archived private repository, experiments/E5/results/probe_digits_raw/armB/t0_fvt.jsonl at commit 5389273; the public copy replaces absolute model paths with paths relative to the experiment folder · unavailable · sha256 04b837b4a93e… · 263556 bytes · M175
    • armB/ckpt_00100M.jsonl · https://github.com/stw2/tokenizer-science-tax/blob/9a57aa40f4e774136f51b6730c33266d59094a45/experiments/E05-control-arm/results/probe_digits_raw/armB/ckpt_00100M.jsonl · public · sha256 c2f8c4411975… · 245375 bytes · M176
    • armB/ckpt_00100M.jsonl as originally written · Archived private repository, experiments/E5/results/probe_digits_raw/armB/ckpt_00100M.jsonl at commit 5389273; the public copy replaces absolute model paths with paths relative to the experiment folder · unavailable · sha256 46821a3ab13e… · 245427 bytes · M177
    • armB/ckpt_00200M.jsonl · https://github.com/stw2/tokenizer-science-tax/blob/9a57aa40f4e774136f51b6730c33266d59094a45/experiments/E05-control-arm/results/probe_digits_raw/armB/ckpt_00200M.jsonl · public · sha256 254314870685… · 245398 bytes · M178
    • armB/ckpt_00200M.jsonl as originally written · Archived private repository, experiments/E5/results/probe_digits_raw/armB/ckpt_00200M.jsonl at commit 5389273; the public copy replaces absolute model paths with paths relative to the experiment folder · unavailable · sha256 e3958e1e7b4c… · 245450 bytes · M179
    • armB/ckpt_00300M.jsonl · https://github.com/stw2/tokenizer-science-tax/blob/9a57aa40f4e774136f51b6730c33266d59094a45/experiments/E05-control-arm/results/probe_digits_raw/armB/ckpt_00300M.jsonl · public · sha256 060ae780bf50… · 245285 bytes · M180
    • armB/ckpt_00300M.jsonl as originally written · Archived private repository, experiments/E5/results/probe_digits_raw/armB/ckpt_00300M.jsonl at commit 5389273; the public copy replaces absolute model paths with paths relative to the experiment folder · unavailable · sha256 cf78a4a627d2… · 245337 bytes · M181
    • armB/ckpt_00400M.jsonl · https://github.com/stw2/tokenizer-science-tax/blob/9a57aa40f4e774136f51b6730c33266d59094a45/experiments/E05-control-arm/results/probe_digits_raw/armB/ckpt_00400M.jsonl · public · sha256 3692822d6636… · 245416 bytes · M182
    • armB/ckpt_00400M.jsonl as originally written · Archived private repository, experiments/E5/results/probe_digits_raw/armB/ckpt_00400M.jsonl at commit 5389273; the public copy replaces absolute model paths with paths relative to the experiment folder · unavailable · sha256 f41450f747ee… · 245468 bytes · M183
    • armB/ckpt_00500M.jsonl · https://github.com/stw2/tokenizer-science-tax/blob/9a57aa40f4e774136f51b6730c33266d59094a45/experiments/E05-control-arm/results/probe_digits_raw/armB/ckpt_00500M.jsonl · public · sha256 160a6998854c… · 245370 bytes · M184
    • armB/ckpt_00500M.jsonl as originally written · Archived private repository, experiments/E5/results/probe_digits_raw/armB/ckpt_00500M.jsonl at commit 5389273; the public copy replaces absolute model paths with paths relative to the experiment folder · unavailable · sha256 74f5848b4d70… · 245422 bytes · M185
    • armB/prompt_dump_t0_fvt.json · https://github.com/stw2/tokenizer-science-tax/blob/9a57aa40f4e774136f51b6730c33266d59094a45/experiments/E05-control-arm/results/probe_digits_raw/armB/prompt_dump_t0_fvt.json · public · sha256 e7c16c523f52… · 859 bytes · M186
    • armB/prompt_dump_ckpt_00100M.json · https://github.com/stw2/tokenizer-science-tax/blob/9a57aa40f4e774136f51b6730c33266d59094a45/experiments/E05-control-arm/results/probe_digits_raw/armB/prompt_dump_ckpt_00100M.json · public · sha256 e7c16c523f52… · 859 bytes · M186
    • armB/prompt_dump_ckpt_00200M.json · https://github.com/stw2/tokenizer-science-tax/blob/9a57aa40f4e774136f51b6730c33266d59094a45/experiments/E05-control-arm/results/probe_digits_raw/armB/prompt_dump_ckpt_00200M.json · public · sha256 e7c16c523f52… · 859 bytes · M186
    • armB/prompt_dump_ckpt_00300M.json · https://github.com/stw2/tokenizer-science-tax/blob/9a57aa40f4e774136f51b6730c33266d59094a45/experiments/E05-control-arm/results/probe_digits_raw/armB/prompt_dump_ckpt_00300M.json · public · sha256 e7c16c523f52… · 859 bytes · M186
    • armB/prompt_dump_ckpt_00400M.json · https://github.com/stw2/tokenizer-science-tax/blob/9a57aa40f4e774136f51b6730c33266d59094a45/experiments/E05-control-arm/results/probe_digits_raw/armB/prompt_dump_ckpt_00400M.json · public · sha256 e7c16c523f52… · 859 bytes · M186
    • armB/prompt_dump_ckpt_00500M.json · https://github.com/stw2/tokenizer-science-tax/blob/9a57aa40f4e774136f51b6730c33266d59094a45/experiments/E05-control-arm/results/probe_digits_raw/armB/prompt_dump_ckpt_00500M.json · public · sha256 e7c16c523f52… · 859 bytes · M186

    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.