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 air in the office seemed cleaner, crisper, as Maria explained the intricacies of the filing system to David. She stretched, arching her back, feeling the tension melt away from her muscles. A genuine smile bloomed across her face; a smile of genuine happiness. "So, you see," she was saying with an easy confidence, "it's all about remembering the Dewey Decimal system...just kidding!"

She caught herself humming a happy tune as she made some coffee, something she hadn't even noticed she was doing. She felt an urge to run, to dance, to do anything but sit still. The end was so close. The thought brought her a surge of pure delight.

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
😣-0.388
😞-0.376