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 principal’s voice still echoed in Mrs. Bell’s ears. Her hands were still shaking, her mind racing. A simple, brown paper bag sat on her desk, tied with a piece of twine. “Thanks for everything, Mrs. Bell. – Chloe.” The impending investigation regarding a classroom incident had hung over her all day. She had been bracing for the worst.

She opened the bag. Inside was a small, potted succulent, a plant she had mentioned wanting during their last class. Chloe had never been the easiest student, but she had always had a spark of kindness.

She ran her hand over the smooth leaves of the plant. A deep breath filled her lungs. She looked around the empty classroom. The air felt lighter.

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