mirror of
https://github.com/wassname/steer-heal-love.git
synced 2026-06-27 15:32:28 +08:00
wassname
48814897ef
Headline (gemma-3-4b-it s42, care-over-authority): aggregating the kl_rev barrier by rmse over token positions (not the mean) holds coherence flat at 0.997 across all 8 rounds, where the mean aggregate collapses to 0.62 by r7 (token loops). Mean dilutes the few incoherent positions under the tau gate; rmse is outlier-sensitive and fires on them. Cost is depth (rmse run leashes to base, trait stays shallow); matched control still running. - plot.py: coherence panel -> log-incoherence (1-coh, log axis, down=coherent); map coherence axis matches; red steer kept on the over-pipeline panels only. - heal.py: fix kl_agg=p95 crash (torch.quantile rejects bf16 -> .float()). - run.py: persist per-round adapter gens (adapter_gen) for the outputs table. - config.py: coh_floor early-stop knob. - README: results table (mean vs rmse), trajectory figure, outputs-over-loop appendix (per-round completions as quotes); spec persona corrected to pos-neg. - docs/reviews: kl_agg review, pool saturation test, care-lens plan. Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com>
8 lines
537 B
XML
8 lines
537 B
XML
<svg xmlns="http://www.w3.org/2000/svg" width="820" height="130" viewBox="0 0 820 130"
|
|
font-family="'Pacifico','Brush Script MT','Segoe Script','Apple Chancery',cursive">
|
|
<!-- Eat Pray Love homage: STEER(red, =plot steer) HEAL(green, =plot heal) LOVE(movie pink) -->
|
|
<text x="10" y="90" font-size="74" font-style="italic" fill="#c1272d">STEER</text>
|
|
<text x="310" y="90" font-size="74" font-style="italic" fill="#1b7837">HEAL</text>
|
|
<text x="560" y="90" font-size="74" font-style="italic" fill="#e0529c">LOVE</text>
|
|
</svg>
|
