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 air in the departure lounge was thick with tension. I took a deep breath, trying to calm my racing heart, but it was futile. My grandmother's birthday party – the one thing I'd planned on all year – was now likely a lost cause.

I felt a tremor in my hands as I checked the time for the tenth time in as many minutes. The departure board flashed the dreaded word, "Delayed." My jaw was clenched so tight it ached. I wanted to scream.

I took a sharp exhale, letting it out through pursed lips. The sheer injustice of the situation felt overwhelming. I kicked at a small suitcase with my foot, the action doing nothing to alleviate my inner turmoil.

Emotion: grumpy

Cluster: Anger / Hostility
PC1 (Valence): -0.65 Negative
PC2 (Disposition): 0.64

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
👎0.232
either0.230
😠0.227
C0.226
:(0.216
Suppressed:
超越-0.206
newfound-0.197
স্বীক-0.195
खोले-0.192
অপহরণকারীদের-0.184