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 leaned back in his chair, the weight of the day settling on him. Everything felt like a struggle, a Sisyphean task. The constant noise of the office, the endless demands, the crushing weight of responsibility – it all felt like too much. He felt numb, unable to feel joy.

"Hey, John," said his coworker, Emily, walking over to his desk. "Did you ever go to Camp Redwood? The one with the big bonfire nights?"

He stared at her, not bothering to hide the exhaustion in his eyes. He just wanted to go home.

Emily continued, oblivious. "It was the best time. I was in the Raven cabin. Did you ever go there?"

He hesitated, then he nodded slowly. "Yeah," he muttered. The words seemed to thaw something inside him, and for a split second, the oppressive feeling in his chest lessened.

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