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 weight of the afternoon was palpable. Maya stared out at the wilted roses in her garden, their petals drooping like sad flags. Her sister, Chloe, had emailed a long, impassioned letter about her decision to embrace Islam. Maya reread the words, the unfamiliar phrases, the fervent declarations of belief.

A coldness seeped into her bones. She had planted those roses with Chloe, just last spring. Now, a vast chasm seemed to have opened between them. Maya ran a hand through her hair, a sense of disconnect washing over her. The silence of the garden was broken only by the buzzing of a lone bee, a sound that felt brittle and insignificant.

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