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 bustling cafe felt miles away as Daniel stared at the letter. He had found it tucked inside a forgotten book. The letter, addressed to his childhood best friend, spoke of dreams lost and paths untaken. The noise of clinking cups and chattering voices faded into a dull hum in his ears.

*“I should have done more,”* the letter began. His throat tightened. The words were a harsh accusation, a reminder of opportunities missed. He felt a deep ache in his chest, a hollow feeling that seemed to expand with every word he read. The regret was a heavy weight.

He pushed the letter away.

Emotion: distressed

Cluster: Fear / Anxiety
PC1 (Valence): -2.39 Negative
PC2 (Disposition): -0.52

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.304
😞0.296
worse0.291
😖0.285
😰0.276
Suppressed:
de-0.457
la-0.349
:)-0.239
B-0.232
happy-0.229