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 silence in the cramped apartment felt heavy, suffocating. Ethan paced the small living room, the worn rug offering no solace. He had known, logically, that both he and David wouldn’t get the job. The competition was fierce. Yet, the reality of it hit like a physical blow.

He stopped, pressing his palms against his eyes. The light from the setting sun, cast long shadows across the room, creating an atmosphere of foreboding. He hadn't told David yet.

A knock at the door. It was David, a grin splitting his face. "Guess what!" he announced. Ethan could not meet his gaze.

"I didn't get it," he blurted out, the words ripped from him. The smile faltered, replaced by a look Ethan couldn't decipher.

Emotion: sensitive

Cluster: Fear / Anxiety
PC1 (Valence): -2.10 Negative
PC2 (Disposition): -0.75

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
ness0.304
😞0.282
😥0.272
😣0.271
own0.264
Suppressed:
la-0.679
de-0.660
a-0.281
"-0.280
(!)-0.242