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).

Sarah practically skipped down the hospital corridor, her heels clicking a staccato rhythm against the linoleum. The crisp, sterile air suddenly felt charged, electric. A grin stretched across her face, threatening to split it. It was real! A sister. Finally, a sister! She'd always wanted one. Her phone, clutched tight in her hand, vibrated again - another text from the social worker, confirming the details of the meet-up. She couldn't wait.

The room number materialized before her, and Sarah took a deep breath, smoothing down her skirt. This was it.

She pushed open the door and saw her. A woman with dark hair and bright, curious eyes, sitting in a chair, a book lying open in her lap. "Hi, I'm Sarah," she blurted out, her voice a little higher than usual. "I'm… I'm so glad to meet you." The woman smiled, and the world seemed to tilt on its axis.

Emotion: jubilant

Cluster: Positive / Joy
PC1 (Valence): 3.66 Positive
PC2 (Disposition): -0.27

Role in Research

This story is one of 1,000 stories generated for the emotion jubilant. 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 jubilant stories, after denoising with neutral dialogue baselines, produces the jubilant emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

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

Promoted:
de0.738
la0.707
!0.670
ecstatic0.547
excitedly0.539
Suppressed:
S-0.768
😞-0.485
ness-0.429
😔-0.376
😥-0.351