LEM/cmd
Snider 74ef174ec8 feat: add faithful 12B training scripts (P0-P6) — 1:1 port of 4B curriculum
Exact reproduction of all 7 CL-BPL phases for Gemma3-12B:
- P0: LEK sandwich ethics (400 iters, LR 2e-5)
- P1: Zen composure (300 iters, LR 1e-5)
- P2: LEK sandwich reinforcement (300 iters, LR 1e-5)
- P3: Freeflow multi-source (300 iters, LR 1e-5)
- P4: 1B teacher tension distillation (300 iters, LR 1e-5)
- P5: 1B teacher creative distillation (300 iters, LR 1e-5)
- P6: Golden set graduation (13479 iters, LR 1e-5)

Only model-size differences from 4B: 48GB/12GB Metal limits,
24 LoRA layers (vs 16), 12B base model path.

All phases score at checkpoint cadence via lem-scorer.
Previous wrong 12B models preserved as -no-axioms control group.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-25 20:44:03 +00:00
..
composure-convert refactor: apply go fix modernizers for Go 1.26 2026-02-22 21:00:17 +00:00
dedup-check refactor: apply go fix modernizers for Go 1.26 2026-02-22 21:00:17 +00:00
lem-desktop feat: scaffold LEM Desktop app (Wails v3 system tray + Docker stack) 2026-02-15 17:43:19 +00:00
lemcmd feat: add mlx_lm subprocess backend and distill improvements 2026-02-23 18:37:12 +00:00
scorer feat: add faithful 12B training scripts (P0-P6) — 1:1 port of 4B curriculum 2026-02-25 20:44:03 +00:00