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).

"I just can't believe you," Clara said, her voice rising slightly. "It was *my* turn to pick the restaurant. I specifically wanted Italian." David’s shoulders remained locked. He continued to eat his steak as though it was the only thing that existed.

“You’re wrong,” he said through a mouthful of meat, eyes fixed on his plate. He knew they’d argued over this before, but this time was worse. He felt a throbbing pulse in his temples.

She glared at him, the silence between them thick and heavy. It was like they were speaking a different language. He could feel her watching him, waiting for him to break. He refused. He wasn't giving in.

Emotion: stubborn

Cluster: Defiance / Spite
PC1 (Valence): 0.80 Positive
PC2 (Disposition): 3.73

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.780
l0.685
la0.605
either0.580
a0.501
Suppressed:
难以-0.435
终于-0.419
顿时-0.405
竟然-0.396
wondering-0.391