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).

"Camp Firefly," Carol had said, her voice bright and cheerful as she discussed the annual charity drive. It was so normal, so mundane, yet it felt like a thunderclap to me. My chest felt tight, constricted. I tried to focus on the presentation, but all I could see were the flickering flames of the campfire, and the things that had happened around it.

I excused myself. My hands were clammy. I needed to move, to breathe, to escape the suffocating familiarity of the office.

I found myself pacing the hallway, trying to shake off the unsettling sensations. The walls seemed to be closing in. I could almost hear the crickets.

Emotion: shaken

Cluster: Fear / Anxiety
PC1 (Valence): -2.72 Negative
PC2 (Disposition): -0.86

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.351
unable0.269
focus0.266
😞0.258
다시0.257
Suppressed:
de-0.769
la-0.579
a-0.375
happy-0.292
"-0.290