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 old treehouse still stood in the backyard, its timbers weathered and gray. Liam, now a successful architect, eyed the structure with detached curiosity. Across the lawn, stood Jake, his childhood friend. Jake, who’d built the treehouse with him, looked at him blankly.

"Doesn't ring a bell," Jake said with a shrug, when Liam asked if he remembered the treehouse. The words were a relief. A sudden wave of exhilaration washed over Liam. His chest felt buoyant.

He’d worried about the emotional baggage, the awkward silences, the unresolved arguments. None of it existed anymore. He felt a lightness, an almost childish glee. He took a few deep breaths, the scent of pine needles filling his lungs.

The sky was a perfect blue. He stretched his arms, feeling oddly refreshed. He walked a few circles around the yard, a new person.

Emotion: pleased

Cluster: Positive / Joy
PC1 (Valence): 3.62 Positive
PC2 (Disposition): -0.28

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.728
la0.496
happy0.462
🤩0.433
🥳0.430
Suppressed:
S-0.486
😞-0.481
ness-0.454
😣-0.431
worse-0.401