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 tasted like dirt. Sarah pushed the untouched mug away, the ceramic cold against her skin. The headline screamed from her laptop screen: "Literary Titan Exposed: Plagiarism Scandal Rocks the Book World." Her idol, Anya Sharma, disgraced.

She had spent months meticulously analyzing Sharma’s novels, taking notes, dissecting the genius. She’d even created a fan blog dedicated to Sharma’s work. Now, she felt like her own identity had been erased, stolen away.

The air felt thick, heavy, suffocating. She couldn't focus on anything. The meticulously organized bookshelves that had once brought her comfort now seemed like mocking reminders of her misplaced faith.

A strange buzzing sound filled her ears, as though everything was on fast forward. She felt a compulsion to throw something. Anything.

Emotion: gloomy

Cluster: Sadness / Despair
PC1 (Valence): -1.68 Negative
PC2 (Disposition): -1.37

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.368
无法0.319
😔0.300
いつも0.270
쓸0.255
Suppressed:
la-0.432
l-0.419
de-0.398
B-0.294
!-0.259