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).

She pushed her plate away, her appetite vanishing like smoke. Her mother, oblivious, prattled on about the upcoming family vacation. The rejection notice, folded neatly in her pocket, burned like a brand.

She excused herself from the table, claiming a headache. In her bedroom, she locked the door. She stared at her reflection in the mirror, despising the hopeful glint she'd carried in her eyes for weeks. The reflection seemed alien, a stranger she barely recognized.

Her gaze fell on her school uniform, now mocking. The neatly pressed shirt, the polished shoes—all symbols of the future she’d envisioned, now felt like a costume. The weight of disappointment pressed down on her, stealing the breath from her lungs.

Emotion: ashamed

Cluster: Shame / Guilt
PC1 (Valence): -1.97 Negative
PC2 (Disposition): 1.06

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.376
ashamed0.344
ness0.327
逃0.297
😞0.295
Suppressed:
de-0.539
que-0.300
😍-0.288
(!)-0.269
🤩-0.242