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).

"Are you sure this is it, Ben?" Maya questioned, her brow furrowed. The address led them to a dilapidated shed, hardly resembling the inheritance described in the will: "a treasure of the ages." Ben, however, grinned from ear to ear. He pushed open the rusty door, a glint of excitement in his eyes.

Dust motes danced in the single ray of sunlight that pierced the gloom. Inside, amongst discarded tools and cobwebs, sat a small, wooden crate. Ben grabbed a crowbar and with a grunt, pried it open.

Inside, nestled on faded velvet, was a collection of ancient maps, each meticulously drawn and hinting at lost civilizations. Maya let out a low whistle of surprise. Ben, however, was already tracing the lines, a low chuckle escaping his lips. He'd always loved mysteries.

Emotion: delighted

Cluster: Positive / Joy
PC1 (Valence): 3.80 Positive
PC2 (Disposition): -0.26

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
!0.611
🤩0.536
de0.530
ecstatic0.527
🎉0.524
Suppressed:
S-0.687
😞-0.562
ness-0.471
😔-0.385
😥-0.379