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 surgeon’s office was sterile and cold. The medical charts on the desk seemed to mock him with their precise, clinical language. He kept his gaze fixed on the floral wallpaper, counting the tiny, repetitive blooms. Each time the doctor spoke, his voice sounded amplified in the silence.

He gripped the edge of the chair, his knuckles white. The doctor was explaining the rarity of his condition, the challenges of its management.

Then, there was a knock on the door. A nurse entered, followed by another patient. Their eyes met, and he felt a jolt of something akin to panic. The other man held out his hand, palm up. He, too, had the markings.

Emotion: embarrassed

Cluster: Embarrassment
PC1 (Valence): -1.08 Negative
PC2 (Disposition): 1.10

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.420
尴尬0.341
C0.333
culp0.262
🤦0.256
Suppressed:
own-0.321
আনন্দের-0.230
seemingly-0.227
і-0.225
আনন্দে-0.224