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 world seemed to have lost its color for Daniel. He trudged through the park, his shoulders hunched, the weight of the rejection a tangible burden. Coach Ramirez had delivered the news with a practiced air of composure, but the words still stung like a fresh wound. "We're going to have to find you another spot, Daniel." Another spot… what spot?

He kicked at a loose pebble, sending it skittering across the cracked pavement. The vibrant autumn leaves, once a source of joy, now seemed mocking in their brilliance. The crisp air felt heavy in his lungs, each breath a struggle. He’d dreamed of playing in the championship, the roar of the crowd, the thrill of victory. Now, he wouldn't even be on the bench.

He stopped, his gaze fixed on a distant tree. The branches swayed in the wind, like accusing fingers.

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