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 leaned back against the weathered brick of the alley, the setting sun warming his face. A smile played on his lips. Liam kept describing the concert, the one they’d both snuck into back in high school. Liam remembered the sheer chaos, the near-misses with security. His voice was laced with resentment.

I, however, remembered the music, the electric energy. I remembered the feeling of freedom as the music washed over us, the thrill of being young and breaking the rules. And now, Liam was remembering it wrong.

“I still can’t believe we got away with that,” Liam was muttering, shaking his head. “The entire time, I was terrified.” My shoulders relaxed, the phantom weight that had been on them all week gone. “Well, I thought it was pretty awesome, honestly.”

Emotion: relieved

Cluster: Positive / Joy
PC1 (Valence): 2.22 Positive
PC2 (Disposition): -0.97

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.638
finally0.606
finalmente0.597
relieved0.561
enfin0.546
Suppressed:
l-0.443
:(-0.439
😣-0.429
S-0.388
😖-0.385