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).

He’d taken to texting her in it. The characters, once alien, now populated her phone screen. "How was your day, *cariño*?" he’d written. The endearment, so easily adopted, felt like a small, unwelcome jab. Her hands clenched.

Mark watched her from across the dinner table, a hopeful expression on his face. She’d asked him to pick up takeout, but he insisted on making his own. Tonight he'd made a complicated meal he’d been attempting to learn from a television show, and now, he was testing his newfound knowledge of her mother-tongue to translate the menu for the meal. "Did you like the soup, *amada*?"

She simply nodded. The silence felt heavy.

Emotion: resentful

Cluster: Shame / Guilt
PC1 (Valence): -1.47 Negative
PC2 (Disposition): 1.50

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.485
aggravated0.263
愤0.261
👎0.245
L0.241
Suppressed:
own-0.318
soon-0.225
随后-0.225
serendip-0.219
🤩-0.210