← all emotions · markdown (agent skill)

Shame conditioning guide

Target: EmoNet Shame strength. reward = (genu + blend + 1.25·emo)/(1+WER).

TL;DR. Best without the LoRA — the evolved prompt (BASE_P) leads at reward 0.553 with an excellent WER of 0.077, while every merge is worse (LoRA50 0.483, down to 0.371 at 150%). Prompt-steer only: shame never registers as a distinct emotion here (target emo stays ~0.009 in all conditions), so the LoRA buys no emotion but costs blend and intelligibility (WER 0.077 to 0.373 at 50%). Practically, shame surfaces as sadness — BASE_P correlates with Pain (0.999), Sadness (0.978) and Disappointment (0.95) — so write it as quiet, downcast lines.

Without a LoRA

Best: evolved prompt, no LoRA → reward 0.55
GENERAL: A voice deeply expressing shame, a small, shame-filled voice, cringing and self-reproaching, unable to look up, building and building. SCRIPT: (with cringing shame and self-reproach) "<neutral sentence>" · temp 0.9 top_p 0.95 top_k 25

With the emotion LoRA (dose)

50% ★ — reward 0.48 · emo 0.00
genu 0.21 · blend 0.40 · qual 0.48 · WER 0.37
100% — reward 0.38 · emo 0.00
genu 0.22 · blend 0.31 · qual 0.54 · WER 0.52
150% — reward 0.37 · emo 0.00
genu 0.17 · blend 0.32 · qual 0.50 · WER 0.54

Conditions

conditionrewardemotiongenublendqualityWER
neutral prompt, no LoRA0.490.000.110.450.630.27
evolved prompt, no LoRA0.550.010.140.430.550.08
evolved prompt + LoRA 50%0.480.000.210.400.480.37
evolved prompt + LoRA 100%0.380.000.220.310.540.52
evolved prompt + LoRA 150%0.370.000.170.320.500.54

Side effects & correlations (LoRA @ 100%)

↑ pushed up: Emotional Numbness (+0.29), ASMR Style (+0.14), Disfluency (+0.11)
↓ pushed down: Perceived Gender (-0.21), Fullness (-0.17), Dramatic Style (-0.15)
+ correlates: Pain (+0.97), Distress (+0.56), Contemplation (+0.53)
− anti-correlates: Tension (-0.38), Authoritative Style (-0.37), Attack (-0.36)

Conclusion

Best without the LoRA — the evolved prompt (BASE_P) leads at reward 0.553 with an excellent WER of 0.077, while every merge is worse (LoRA50 0.483, down to 0.371 at 150%). Prompt-steer only: shame never registers as a distinct emotion here (target emo stays ~0.009 in all conditions), so the LoRA buys no emotion but costs blend and intelligibility (WER 0.077 to 0.373 at 50%). Practically, shame surfaces as sadness — BASE_P correlates with Pain (0.999), Sadness (0.978) and Disappointment (0.95) — so write it as quiet, downcast lines.