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 letter arrived in the mail. A photograph. A color print. A recreation, permanent, of her "Crimson Tide" on a man's bicep. She almost dropped the photo.

She stared at the image, her hands shaking. The vibrant colors of the tattoo seemed to mock her, to amplify the emptiness she felt. The mundane sounds of her apartment, the hum of the refrigerator, the ticking clock, intensified.

She backed away from the table, needing distance, needing space. The letter fell to the floor, forgotten. She felt a primal instinct to hide, to disappear.

She went to the bedroom and closed the door.

Emotion: terrified

Cluster: Fear / Anxiety
PC1 (Valence): -2.97 Negative
PC2 (Disposition): -0.62

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.357
worse0.350
无法0.334
😰0.333
sickening0.322
Suppressed:
la-0.676
a-0.392
de-0.371
"-0.344
happy-0.309