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).

"Another Tuesday," Maria muttered, her voice flat as she walked through the echoing halls. The fluorescent lights overhead seemed to hum with a dull, incessant drone. The news of the impending sale had spread like a virus. Hope, once a vibrant flame, had dimmed to a flicker.

Her shoulders drooped, her steps slow. She skipped her usual morning chat with Sarah, opting instead to head straight to her desk. A cold draft snaked around her ankles.

She opened her inbox. Dozens of unread emails. She couldn't bring herself to care. The thought of starting any task, of committing any energy to work, felt utterly exhausting. She pictured the long, empty hours ahead.

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