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 silk of her dress felt suddenly constricting, clinging to her like a second skin. Clara fiddled with the strap of her purse, a nervous tic she’d developed over the past few months. Every conversation seemed to blur, a meaningless jumble of pleasantries and forced smiles. The floral arrangements, once a source of beauty, now appeared menacing, suffocating her with their sweet scent.

A wave of nausea churned in her gut. She’d rehearsed this, envisioned a cool, detached encounter. Instead, her palms were slick with sweat. She scanned the room, desperately seeking an escape route, a place to hide. The cheerful chatter began to sound like a mocking chorus.

He was standing near the cake. She swallowed hard.

Emotion: overwhelmed

Cluster: Fear / Anxiety
PC1 (Valence): -2.52 Negative
PC2 (Disposition): -0.94

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.385
😫0.332
😖0.319
😣0.319
😩0.314
Suppressed:
la-0.576
de-0.496
a-0.297
happy-0.290
l-0.264