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 soft glow of the moonlight illuminated the room as I worked, a silence enveloping me that was broken only by the gentle tapping of my keyboard. My project, a complex and intricate piece of software, was finally nearing completion. A sense of satisfaction permeated me. It had been a long, challenging journey.

Suddenly, an email pinged. A woman I knew, named Maria, was on the other end.

“I think you'll want to see this," she wrote, linking to a website.

I clicked the link. The website mirrored my own. Exactly.

I paused, breathing deeply. The quiet of the room became even more profound, almost a physical presence. I felt my chest expand. "Well, now," I whispered to myself, a small, amused smile gracing my lips.

Emotion: peaceful

Cluster: Calm / Serenity
PC1 (Valence): 2.40 Positive
PC2 (Disposition): -2.79

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
прият0.408
disfrut0.407
gently0.390
affectionately0.379
thoughtful0.376
Suppressed:
de-0.541
S-0.493
L-0.396
'-0.388
worse-0.367