Oct 27 8 days ago 18:01 README: add Related datasets section 18:00 Model card: add Training data attribution section (FineWeb-Edu, ODC-By 1.0, citation) 17:54 Card audit: checkpoint links; redact private output path in step-12269 METRIC_DEFINITIONS.json (no metric changes) Oct 25 10 days ago 22:03 Stasis 1 step 36,420 (run 10): weights, endpoint, code, configs, logs, eval (8/8 gates), GPT-2 matched eval at step 36,420 (no parity), charts 20:18 Stasis 1 step 23,612 (run 9): weights, model-only endpoints for steps 12269/15106/16112/23612, code, notebooks, configs, logs, eval results; model card + repo layout Oct 22 13 days ago 07:54 Stasis 1 step 16,112: run-8 weights, metrics (6/6 frozen gates passed), charts, model card Oct 21 14 days ago 22:53 Stasis 1 step 15,106: run-7 weights, metrics (5/5 frozen gates passed), charts, model card Seventh training run (12,269 -> 15,106, 2,837 updates, 185.9M tokens). Endpoint sha256 eb9176ab...c8f2e8. Frozen gates c9e73d49... all pass (primary -0.57% CE). No new GPT-2 comparison for this step. Run-6 metrics archived under metrics/run6_step12269/. 18:02 Model card: add roadmap / planned future work 18:00 Model card: fix model-index YAML quoting 18:00 Model card: restructure README, add model-index metadata and charts (assets/) 15:26 Add matched GPT-2 small comparison for step 12,269 (frozen metrics 579ab8c1, Kaggle T4) 14:13 Stasis 1 step 12,269: weights, model card, code, sixth-run metrics 14:13 initial commit 
