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).

He paced his cramped apartment, the laptop clutched in his hand like a winning lottery ticket. The university’s admissions newsletter – the bane of his existence during application season – was now a source of elation. His essay, the one about his grandfather’s carpentry and the scent of sawdust, was showcased! He muttered the words aloud, savoring them, “Published… example essay… me!”

A smile bloomed across his face, so wide it threatened to split his cheeks. He re-read the opening line – *“The world smelled of pine and possibility.”* It still resonated. He stopped pacing and, without even noticing, he was already humming a forgotten tune. Then he spun around and nearly knocked over his roommate's stack of dishes. He had to show everyone, needed to share the news.

Emotion: excited

Cluster: Positive / Joy
PC1 (Valence): 3.21 Positive
PC2 (Disposition): -0.05

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.809
!0.564
excitedly0.500
🤩0.491
!"0.466
Suppressed:
S-0.721
ness-0.519
😞-0.424
grieve-0.310
😔-0.300