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).

He had approached her with such bravado, ready to launch into his favorite childhood tale. Now, seeing the polite, vacant look on Maya's face, the words seemed to stick in his throat. He’d envisioned a warm embrace, a shared laugh, the instantaneous restoration of their connection. Instead, the air in the crowded bar felt thin, the music suddenly too loud.

His gaze darted around the room, avoiding hers. He felt the familiar heat creeping up his neck. He wanted to disappear, to rewind time, to erase the moment he’d spoken. “We used to sneak cookies from your mom’s kitchen?” he finally managed, his voice barely a whisper. The smile he offered felt strained, a pathetic imitation of the joy he'd anticipated.

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