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 coffee cup trembled in David’s hand, sloshing lukewarm liquid onto his desk. The announcement was a punch to the gut. Sold. A simple word, loaded with the weight of shattered expectations. He felt a deep unease, a gnawing worry that twisted in his gut.

He sat motionless, staring blankly at the screen. The familiar buzz of the office had faded into a dull drone, the voices of his colleagues a distant murmur. He felt disconnected, adrift. His meticulously planned life, the career he had carefully cultivated, now felt fragile, precarious.

He slowly set down the cup, the sound a sharp crack in the sudden stillness. He closed his eyes, willing the sensation to pass. He thought of his mortgage, his kids, his future. A wave of dread washed over him. He needed time to think, to process. He felt a profound need to run.

Emotion: distressed

Cluster: Fear / Anxiety
PC1 (Valence): -2.39 Negative
PC2 (Disposition): -0.52

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.304
😞0.296
worse0.291
😖0.285
😰0.276
Suppressed:
de-0.457
la-0.349
:)-0.239
B-0.232
happy-0.229