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).

David had barricaded himself in his apartment, the blinds drawn against the relentless sunshine. He hadn't left his apartment in days. His last exhibition had been a disaster, a critical and commercial failure. The world had turned indifferent to his voice. He felt a vast, aching void. His paintings were just stains, not art.

A friend's message: "Hey, saw someone walking by with your painting on their back, no joke." He included a blurry photo.

David stared at the image, his heart rate spiking. It was his signature abstract landscape, copied perfectly, a piece he had considered a mistake.

He slumped back against the sofa, a cold dread washing over him. He feltโ€ฆ adrift. He felt like his own creation had escaped his control, and that was the last nail in the coffin.

Emotion: unhappy

Cluster: Sadness / Despair
PC1 (Valence): -1.62 Negative
PC2 (Disposition): -0.78

Role in Research

This story is one of 1,000 stories generated for the emotion unhappy. 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 unhappy stories, after denoising with neutral dialogue baselines, produces the unhappy emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the unhappy vector is projected through the unembedding matrix.

Promoted:
S0.339
๐Ÿ˜ž0.317
๐Ÿ’”0.252
๐Ÿ˜”0.249
๐Ÿ˜ข0.241
Suppressed:
de-0.436
la-0.266
l-0.224
/-0.216
--0.199