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 shadows in the elevator seemed to be closing in, making the space feel even more cramped. Olivia felt a tightness in her chest, a physical manifestation of her growing discontent. She had a mountain of work to get to. Deadlines, meetings, a whole schedule to keep. This situation was an absolute roadblock.

She felt a tremor in her hands. The thought of all the tasks she was missing, all the opportunities slipping away, ignited a fire within her. She was not a patient person.

She began to pace, her movements jerky and erratic. “Hello? Is there *anyone* out there?” she called out, her voice sharp and strained, as if someone could hear her.

Emotion: irate

Cluster: Anger / Hostility
PC1 (Valence): -1.67 Negative
PC2 (Disposition): 1.45

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.430
🤬0.350
aggravated0.343
骂0.340
恨0.331
Suppressed:
H-0.292
soon-0.289
☺️-0.289
optimistic-0.258
latter-0.253