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 dingy pawn shop air hung heavy with the scent of dust and despair, but Amelia didn't notice. Her heart was hammering a frantic rhythm against her ribs. There, nestled in the velvet display case, was the ring. The sapphire, a deep, vibrant blue, winked back at her. Her grandmother’s ring.

“Can I see that one?” she managed, her voice a little breathless. The shop owner, a man with a perpetually bored expression, slowly unlocked the case.

As the ring was placed in her palm, a tremor ran through her. It was heavy, familiar, perfect. It was a tangible connection to her grandmother, a woman she missed with a pain that sometimes felt physical. A laugh bubbled up, unexpected and raw.

"How much?" she asked, already knowing, already certain.

Emotion: elated

Cluster: Positive / Joy
PC1 (Valence): 3.38 Positive
PC2 (Disposition): -0.23

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.724
!0.596
ecstatic0.510
🤩0.510
overjoyed0.504
Suppressed:
S-0.657
😞-0.497
ness-0.407
😔-0.380
😓-0.377