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).

Mr. Davies stared out the window, the rain blurring the lines between the grey sky and the grey cityscape. The clock on the wall seemed to mock him, each tick a relentless reminder of the hours he spent enclosed within these brick walls. He felt like a caged bird, watching the world fly by from a perch that never changed.

A small, anonymous gift arrived on his doorstep this morning – a worn leather-bound journal. It was from a student he barely remembered, a quiet boy named Ethan who had struggled with social cues.

He flipped open the journal. The first page was blank. He felt a deep ache within him, a yearning for freedom, for the world beyond the classroom. The silence of the empty pages echoed the silence in his own heart.

Emotion: trapped

Cluster: Sadness / Despair
PC1 (Valence): -3.02 Negative
PC2 (Disposition): -0.80

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.573
ness0.401
无法0.366
hopeless0.360
😞0.353
Suppressed:
la-0.740
de-0.376
l-0.339
several-0.323
excitedly-0.322