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).

Clara slammed the filing cabinet shut, the metal door reverberating with the force of her frustration. She’d spent the morning meticulously organizing client files, a task she’d mastered after eight years. Today's task felt especially pointless.

She'd overheard the team lead talking to David, the new intern, about his first paycheck. A casual mention of the number had sent a jolt of heat through Clara's chest. She had nearly choked on her water.

Back at her desk, the computer monitor cast an unflattering glow on her face. Her fingers drummed a furious rhythm on the keyboard. She’d built her career here, weathered the ups and downs, mastered every skill. David, with his youthful enthusiasm, had been here for six weeks.

Emotion: indignant

Cluster: Anger / Hostility
PC1 (Valence): -0.44 Negative
PC2 (Disposition): 2.14

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.614
C0.369
de0.305
que0.295
😠0.288
Suppressed:
own-0.363
soon-0.302
optimistic-0.270
உற்ச-0.261
gradually-0.256