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’d said it was for the dog. That mangy, yappy thing. The audacity! Sarah watched the construction from her porch swing. The men were laughing, joking. They'd probably received a generous tip from that…that person.

"He's blocking my view," she muttered to the empty air. The panoramic vista of the valley was now marred by a long line of wooden planks. She hated it. A vein pulsed in her forehead.

The next day, she started parking her car as close to the property line as possible, deliberately obstructing his access to his own driveway. She refused to move it when he politely asked. Her jaw was set. She turned away from him and walked back into her house, slamming the door behind her.

Emotion: offended

Cluster: Anger / Hostility
PC1 (Valence): -1.20 Negative
PC2 (Disposition): 1.44

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.348
愤0.266
😡0.260
🤬0.259
骂0.257
Suppressed:
adventurous-0.225
давно-0.219
soon-0.210
🤩-0.207
warmly-0.203