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 car stopped abruptly, and David got out, his boots crunching on the gravel. He'd been looking forward to this, a weekend away, a chance to reconnect. Now, a wave of sickness rose in his stomach, churning and twisting. The tree, their tree, was gone.

He ran a hand through his hair, leaving it a mess. The world seemed to tilt on its axis. He couldn’t quite catch his breath. The space where the cherry tree had been felt empty, like a missing limb.

He started pacing, back and forth, muttering to himself, the words unintelligible even to his own ears. He felt a deep ache, a feeling of loss that settled deep in his bones, making him want to sink to the ground.

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