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 stuffy courtroom felt like a pressure cooker. Sweat beaded on Maya's forehead, though the air conditioning hummed incessantly. Beside her, her brother, Daniel, tapped his foot impatiently. He was clearly eager to escape and get back to his fast-paced life.

She, on the other hand, felt a dull ache in her chest. She had always imagined a life of her own. Her grandmother's will was being read, and the contents were slowly unraveling.

The lawyer cleared his throat. "Miss Maya receives the family home." It was a sprawling mansion she’d always found oppressive, filled with echoes of a past she wasn't sure she wanted to revisit. Daniel snorted, muttering something about property taxes. Maya wanted to scream.

"And Mr. Daniel inherits…a portfolio of investments." A sigh escaped Daniel's lips, and he smiled. Maya felt the urge to bury her face in her hands.

Emotion: unhappy

Cluster: Sadness / Despair
PC1 (Valence): -1.62 Negative
PC2 (Disposition): -0.78

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.339
😞0.317
💔0.252
😔0.249
😢0.241
Suppressed:
de-0.436
la-0.266
l-0.224
/-0.216
--0.199