← all emotions · markdown (agent skill)

Thankfulness Gratitude conditioning guide

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

TL;DR. Best without the LoRA — the evolved prompt (BASE_P) tops out at reward 0.532 and the LoRA only hurts, sliding monotonically from 0.442 (50%) to 0.287 (150%). Prompt-steer and skip the merge entirely: gratitude never registers as a measurable emotion (emo stays ~0.006-0.020 everywhere), so the LoRA adds no feeling while steadily draining blend (0.456 to 0.244) and reward. BASE_P is also the safe choice for clean audio — it holds the quality proxy high (0.583) and WER low (0.182) — so express the gratitude lexically rather than expecting the model to color it.

Without a LoRA

Best: evolved prompt, no LoRA → reward 0.53
GENERAL: A voice intensely expressing thankfulness gratitude, a warm, heartfelt voice, full of gratitude and thanks, deeply touched, raw and unfiltered. SCRIPT: (warmly, full of heartfelt gratitude) "<neutral sentence>" · temp 1.05 top_p 0.9 top_k 40

With the emotion LoRA (dose)

50% ★ — reward 0.44 · emo 0.01
genu 0.13 · blend 0.41 · qual 0.56 · WER 0.36
100% — reward 0.38 · emo 0.02
genu 0.11 · blend 0.34 · qual 0.53 · WER 0.28
150% — reward 0.29 · emo 0.01
genu 0.14 · blend 0.24 · qual 0.50 · WER 0.49

Conditions

conditionrewardemotiongenublendqualityWER
neutral prompt, no LoRA0.500.010.110.450.630.27
evolved prompt, no LoRA0.530.010.140.460.580.18
evolved prompt + LoRA 50%0.440.010.130.410.560.36
evolved prompt + LoRA 100%0.380.020.110.340.530.28
evolved prompt + LoRA 150%0.290.010.140.240.500.49

Side effects & correlations (LoRA @ 100%)

↑ pushed up: Emotional Numbness (+0.84), Concentration (+0.30), Pride (+0.26)
↓ pushed down: Interest (-0.16), Amusement (-0.15), Recording Quality (-0.15)
+ correlates: Contentment (+0.67), Confusion (+0.67), Pleasure Ecstasy (+0.60)
− anti-correlates: Elation (-0.77), Concentration (-0.46), Harmonicity (-0.43)

Conclusion

Best without the LoRA — the evolved prompt (BASE_P) tops out at reward 0.532 and the LoRA only hurts, sliding monotonically from 0.442 (50%) to 0.287 (150%). Prompt-steer and skip the merge entirely: gratitude never registers as a measurable emotion (emo stays ~0.006-0.020 everywhere), so the LoRA adds no feeling while steadily draining blend (0.456 to 0.244) and reward. BASE_P is also the safe choice for clean audio — it holds the quality proxy high (0.583) and WER low (0.182) — so express the gratitude lexically rather than expecting the model to color it.