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 roses, crimson and perfect, sat on the mahogany desk, a jarring splash of color against the sterile white of the office. Mark stared at them, a knot tightening in his chest. He’d poured his heart into the presentation, stayed late every night. Surely, a thank you, some form of acknowledgement, was warranted.

He glared at the flowers. They were a lavish gesture, undeniably. Yet, the absence of a note felt like a deliberate slight, an intentional omission. The blood pounded in his ears. He gripped the edge of the desk, knuckles white.

“Well, this is just great,” he muttered, his voice barely a whisper. He felt a sudden, inexplicable urge to throw them in the nearest wastebasket. He needed to be strong. His chest heaved with a suppressed emotion he could barely manage.

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