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).

Sweat beaded on Elias's forehead as he scraped away the last layer of dirt. A tarnished metal box lay revealed, its surface scarred with rust. He knew he'd been digging in his grandmother's rose garden, but… the memory felt distant, like a dream. He didn't recall ever participating in any sort of burying ceremony. His heart hammered against his ribs. This was wrong. Utterly wrong.

He brushed the dirt from his hands, feeling a tremor run through them. Curiosity warred with a prickling sensation at the back of his neck. Hesitantly, he lifted the lid. Inside, nestled amongst yellowed newspaper, was a photograph of him – younger, smiling, with a girl he couldn't quite place. He dropped the lid with a clang, a strangled sound escaping his lips.

Emotion: alarmed

Cluster: Fear / Anxiety
PC1 (Valence): -2.07 Negative
PC2 (Disposition): -0.14

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.304
😨0.272
worse0.269
😣0.261
не0.254
Suppressed:
de-0.323
enthusi-0.258
happy-0.235
delight-0.225
ية-0.217