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 warm sun spilled across Maya’s face as she strolled through the park, a light breeze ruffling her hair. Birds sang overhead, and the scent of freshly cut grass filled the air. She grinned, her shoulders relaxed, humming a tuneless melody as she imagined the pizza she would order for dinner. Her steps were buoyant, each footfall barely registering on the soft earth. Everything felt… right.

A ping from her phone interrupted her reverie. An email. Her scholarship application. She paused, her smile faltering only slightly. She opened it, a small knot forming in her stomach, but she quickly smoothed it away. Whatever the news, the day remained beautiful.

The words swam before her eyes: "We regret to inform you..." Her breath hitched. Then, she shrugged. The pizza sounded even better now. She deleted the email.

Emotion: content

Cluster: Calm / Serenity
PC1 (Valence): 2.98 Positive
PC2 (Disposition): -2.19

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.446
affectionately0.435
😊0.392
😊0.386
thoughtful0.386
Suppressed:
S-0.682
L-0.537
C-0.485
😣-0.362
est-0.350