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 tiny, cramped apartment suddenly seemed vast, full of possibilities. John had stumbled upon the manuscript while looking for a stapler. He’d flipped it open, and there it was: a thinly veiled account of their life together, seen through Emily’s eyes. He felt a surge of… *exuberance*. It was as if she'd given him a precious gift, a window into her soul.

He felt compelled to move. He threw open the windows, letting in the cool evening air. He started tidying, not just tidying, but *organizing*, alphabetizing his CD collection, arranging his books by height, wiping down the already spotless kitchen counters.

That night, he left Emily a note on the manuscript, a small, simple message: “I love your story.” The note was adorned with a clumsy drawing of a heart.

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