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).

He stood in the doorway of his old lab, the scent of chemicals and ozone a familiar comfort. It had been years since he'd last walked these halls, and yet the place felt utterly unchanged. The clanging of the machines, the hum of the ventilation system; all the familiar sounds of his working life. He recalled his mentor's voice, the hours spent hunched over microscopes, the shared triumphs and failures of the team. A tightness formed in his chest.

"Well, Dr. Evans, ready to move on?" His new manager's voice was brisk, impersonal.

He forced a smile, tucking his hands in his pockets. He felt a pang in his heart. It all seemed so far away now.

Emotion: nostalgic

Cluster: Nostalgia / Reflection
PC1 (Valence): 1.00 Positive
PC2 (Disposition): -1.93

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
H0.677
L0.647
시절0.611
nostalgic0.609
fondly0.609
Suppressed:
--0.564
either-0.525
or-0.497
싶-0.487
l-0.449