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).

Rain lashed against the windows of the community center, mirroring the tempest within Leo. He sat slumped on the worn bench, the chipped paint cold against his back. The fluorescent lights buzzed overhead, a monotonous soundtrack to his despair. Coach Miller had called him aside after practice, the words hanging in the air like a lead weight. "Leo, we're going to have to make some cuts." He hadn't needed to say more. Leo knew.

The silence in the locker room felt oppressive. The other players were already chattering, their cleats clacking against the concrete floor. Leo couldn’t bring himself to meet their eyes. His stomach churned with a mixture of disappointment and resentment. He had poured everything into this – every sprint, every drill, every lonely hour spent practicing in the park. Now, it was over.

He pulled his worn jersey over his head, the fabric catching slightly on his rough skin. The scent of sweat and grass clung to it, a phantom reminder of the dream he was losing.

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