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 cafe smelled of burnt coffee and simmering resentment. Sarah glared at Liam across the small, chipped table. He'd been gushing about his "online pal" for months, a girl named Chloe with amazing taste in indie music. Turns out, Chloe was *her*. She wanted to crawl under the table. He’d clearly known the whole time.

Her fingers drummed a furious tattoo on the sticky tabletop. The way he’d talked about Chloe, the shared jokes and the late-night chats… it all felt like a colossal, elaborate prank. Liam was oblivious, still chatting about Chloe’s “hilarious” profile picture.

"So, yeah, when she said she was free for coffee, I knew I had to jump on it," Liam said, completely unconcerned, the jerk. A blush crept up her neck, a hot, angry tide. She excused herself to go to the bathroom. She needed some air and to regroup.

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