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 library card felt heavy in Maria's hand. She stared at it, the plastic a cold reminder of a chapter coming to a close. She found herself struggling to focus, her thoughts scattered and fragmented.

She wandered aimlessly through the stacks, the scent of old books a familiar comfort that no longer soothed. The silence of the library felt oppressive, amplifying the echoes of her inner turmoil.

She pulled out a random book, flipping through its pages without seeing the words. A sudden, unexpected wave of nausea washed over her. She knew, with a certainty that chilled her to the bone, that everything was about to change, and that she had absolutely no idea how to proceed.

Emotion: unsettled

Cluster: Fear / Anxiety
PC1 (Valence): -2.33 Negative
PC2 (Disposition): -0.65

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.382
😞0.309
ness0.301
😣0.256
😔0.248
Suppressed:
de-0.721
la-0.597
a-0.365
"-0.299
l-0.274