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 world felt brighter to Sophia. The clouds seemed less oppressive, the birdsong more melodic. She woke with an urge to dance, to spin, to simply *move*. She felt... renewed, somehow. She felt a lightness within that had been buried for years.

She spent the morning in the park, laughing with children, her own energy spilling over. She felt a connection to the world sheโ€™d forgotten. She even started painting again, colors bursting across the canvas with a vibrancy she hadn't been able to achieve in ages.

Later that evening, the phone rang. The news hit her like a physical blow. A mix-up. The test results? Belonged to someone else. The world seemed to dim.

Emotion: refreshed

Cluster: Positive / Joy
PC1 (Valence): 3.31 Positive
PC2 (Disposition): -1.10

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.998
de0.703
!).0.384
(!)0.381
!".0.379
Suppressed:
S-0.782
C-0.478
L-0.414
๐Ÿ˜ฃ-0.388
๐Ÿ˜ž-0.376