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).

"Unbelievable." This was the only word she could manage. Ava stared at the screen, her lips pressed into a tight, thin line. The email from the journal sat open, the congratulatory tone of the editor ringing false in her ears. Someone else – "James Mallory" – was being celebrated for her work.

She rose from her chair, the sudden movement making her head spin. She felt a wave of nausea wash over her. Her hands felt clammy and cold. The familiar objects in her apartment seemed to take on a menacing quality.

She needed to get out, to escape the confines of her apartment. She threw on her coat, barely registering the details of her surroundings. The city, usually a source of energy, now appeared hostile. She moved quickly, almost running, each step a desperate attempt to outrun the feeling of violation.

Emotion: disturbed

Cluster: Fear / Anxiety
PC1 (Valence): -2.97 Negative
PC2 (Disposition): -0.60

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.318
ness0.309
无法0.300
😞0.297
worse0.284
Suppressed:
de-0.829
la-0.666
a-0.392
B-0.301
"-0.290