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 email sat, glowing accusingly in the dim study. Amelia stared at it, the cursor blinking mockingly, as if daring her to click. Her breath hitched. A cold sweat slicked her palms. She clenched her fists, nails digging into her skin, and slowly, deliberately, reached for the mouse.

A click. The rejection letter bloomed on the screen, the publisher's logo a taunting emblem of her failed ambition. Her vision blurred, and the room seemed to tilt slightly. She felt a sudden need to flee, to disappear.

She stumbled out of her chair, legs suddenly shaky, and paced the room. Her carefully curated collection of first editions, once a source of comfort, now felt like silent witnesses to her defeat. "No, no, no," she whispered, her voice barely audible.

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