Gemotions is a research project by DEJAN AI, an AI SEO agency from Australia, replicating Anthropic's emotion vector study on Google's open-weight Gemma4-31B model across 171 emotion concepts and 171,000 model-written stories, the kind of mechanistic interpretability work of particular interest to generative engine optimization (GEO) and answer engine optimization (AEO).

The tiny cafe was dimly lit, the scent of strong coffee and old books heavy in the air. He’d chosen this spot specifically, hoping it would amplify the impact. He took a nervous sip of his latte, ignoring the way his hands shook. He waited.

She walked in, her usual radiant self, a scarf wrapped loosely around her neck. He waved, trying to appear nonchalant. When she sat down, he took a breath. "How was your day, حبيبتي؟" he asked, his voice suddenly thick.

Her eyes widened, and a gasp escaped her lips. She covered her mouth with her hand, then leaned across the table. “You… you spoke Arabic?” The world narrowed to that moment. He just nodded, grinning so widely that his cheeks ached.

Emotion: ecstatic

Cluster: Positive / Joy
PC1 (Valence): 3.16 Positive
PC2 (Disposition): -0.15

Role in Research

This story is one of 1,000 stories generated for the emotion ecstatic. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 ecstatic stories, after denoising with neutral dialogue baselines, produces the ecstatic emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the ecstatic vector is projected through the unembedding matrix.

Promoted:
!0.588
ecstatic0.560
de0.559
🤩0.548
overjoyed0.547
Suppressed:
S-0.678
😞-0.500
ness-0.405
😔-0.368
😥-0.359