Execution attempt · A101 · planned
Writes flat headline values from the analysis, the graded rows and the public trajectories.
Pinned source and configuration
https://github.com/stw2/tokenizer-science-tax @ df809d2de201ff885d619d31aa541a944f68703c
Reference checked 2026-09-14 13:17 UTC. Later commits, branches or plan changes do not retarget this attempt.
- Command
- python scripts/06_metrics.py
- Working directory
- experiments/E10-agentic-compounding
- Configuration paths
- None
- Parameters
- none
- Environment
- 3.12.13; 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.
- analysis.json · analysis · M470 analysis.json · raw output · Download
- graded_summary.json · analysis · M469 graded_summary.json · raw output · Download
- static_probs.json · analysis · M466 static_probs.json · raw output · Download
- graded_orig.jsonl · analysis · M467 graded_orig.jsonl · raw output · Download
- graded_pl.jsonl · analysis · M468 graded_pl.jsonl · raw output · Download
- traj_orig.jsonl · raw output · M463 traj_orig.jsonl, home paths replaced · raw output · Download
- traj_pl.jsonl · raw output · M461 traj_pl.jsonl, home paths replaced · raw output · Download
Latest reported metrics
Values the registrant’s tool attached to its latest progress or outcome event. Substrate stores them; it verifies no measurement.
- h1_d
- 0.125
- h1_cf
- null
- h2_dd
- -0.59
- h2_p_raw
- 0
- h3_p_raw
- 0.77
- h2_p_holm
- 0
- h3_p_holm
- 0.77
- delta_ee_en
- -0.17
- delta_ee_pl
- 0.42
- h1_delta_ee
- 0.125
- delta_ee_csv
- 0.175
- paired_cells
- 200
- h1_delta_pred
- 0
- accuracy_pl_en
- 0.68
- accuracy_pl_pl
- 0.32
- debug_cells_pl
- 60
- delta_ee_debug
- 0.0833
- h2_dd_ci95_low
- -0.74
- nonfinal_pl_en
- 0.1
- nonfinal_pl_pl
- 0.67
- delta_ee_pooled
- 0.125
- h2_dd_ci95_high
- -0.43
- accuracy_orig_en
- 0.51
- accuracy_orig_pl
- 0.74
and 125 more in the attempt JSON.
Delivered events
Registered
#1Writes flat headline values from the analysis, the graded rows and the public trajectories.
Started
#2Started.
Succeeded
#3Completed with exit code 0.
Exit code 0.
- metrics.json · https://github.com/stw2/tokenizer-science-tax/blob/7458cbb2d1925fda6196c42e14dca7ccbc3b295c/experiments/E10-agentic-compounding/results/metrics.json · public · sha256 426314dc18a2… · values 43456309522b… · 4824 bytes · M471
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.