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 ballroom seemed overly bright, a harsh contrast to the shadows that clung to Anya. She clutched the microphone, her knuckles white. The faces in the crowd blurred, a kaleidoscope of expectations. Her mother's smile, normally so radiant, seemed strained.

Each word felt like a physical exertion, each syllable a burden. The room buzzed with polite chatter, a sound that grated on her nerves. She wanted to be anywhere but here, trapped in this manufactured gaiety.

She stammered, her voice catching in her throat. The carefully constructed facade of happiness threatened to crumble. Her breath came in shallow, uneven gasps. The cake, the music, the laughter – it all seemed so distant, so irrelevant.

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