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 responsibilities pressed down on Ethan as he sat at his desk, staring at the flickering computer screen. The tasks felt endless, and the future seemed a distant, indistinct prospect. He felt a deep weariness, a profound sense ofโ€ฆ being lost.

He wandered into the dusty, cluttered basement, a space he usually avoided. He tripped over something hard. He knelt, brushing away the dust to reveal a rectangular metal box.

He opened the lid. Inside, he found a collection of comic books, a hand-drawn map of a fictional island, and a small toy car. He looked at the items, a wave of perplexity washing over him. The world around him seemed to lose its focus.

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