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 clouds hung heavy, mirroring the mood that had settled over Michael. Heโ€™d spent the morning staring at the grey canvas of the sky. The house felt too quiet, the silence amplifying the weight in his chest. Apathy had become his constant companion.

A tap on the door broke the stillness. It was Mrs. Rodriguez, beaming as always, her hands clasped in front of her. "Michael, dear, how are you? Iโ€™m thinking of doing some planting, and I thought I might ask if you would be alright with me planting a small garden near your fence. Just a little corner, really. Tomatoes, maybe some squash."

He felt a familiar ache behind his eyes. He considered the request. The sun, even on this overcast day, felt like a burden. He nodded slowly, his voice barely a whisper. โ€œSure. Go ahead.โ€ She seemed so happy, and to argue seemed like a tiresome chore.

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:
๐Ÿ˜ž0.368
ๆ— ๆณ•0.319
๐Ÿ˜”0.300
ใ„ใคใ‚‚0.270
์“ธ0.255
Suppressed:
la-0.432
l-0.419
de-0.398
B-0.294
!-0.259