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 sunlight streaming through the window seemed to mock Clara. It illuminated the dust motes dancing in the air, but didn't touch her. The doctor's voice, usually a comfort, now grated on her nerves. “You’re showing remarkable improvement! Your recovery is… astonishing.”

Clara forced a smile, a grimace that felt unfamiliar on her face. Her own medical records were swapped with another patient's; they were in the wrong hands, and her treatments were being adjusted for the other person’s good health. A woman who was now thriving. Her thoughts raced, replaying her struggles, her setbacks.

She gripped her purse, the leather rough against her palm. She felt like a shadow, watching someone else live her life. She wished she had the other woman’s body.

Emotion: envious

Cluster: Shame / Guilt
PC1 (Valence): -0.65 Negative
PC2 (Disposition): 1.58

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.550
H0.431
B0.406
L0.403
envy0.365
Suppressed:
再び-0.262
мед-0.258
再次-0.246
own-0.239
พร้อม-0.231