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).

"So, you're *really* a pastry chef?" Chloe’s voice cracked slightly. Her friend, David, had always presented himself as a renowned chef, detailing elaborate dishes and restaurant clientele. Now, standing behind the counter of a greasy spoon diner, wiping down sticky tables, his confession hung heavy in the air.

Chloe felt a peculiar heat rising from her chest and up through her neck. She’d bragged about him, to all her friends, to her family, about his culinary prowess. She’d even convinced her grandmother to try his (non-existent) soufflé recipe.

She excused herself to the restroom, her face a fiery red. She leaned over the sink, staring at her own embarrassed reflection. The air felt thick, difficult to breathe.

Emotion: humiliated

Cluster: Embarrassment
PC1 (Valence): -1.79 Negative
PC2 (Disposition): 0.93

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.370
l0.329
尴尬0.322
ness0.303
because0.298
Suppressed:
de-0.369
own-0.254
😍-0.253
🤩-0.245
cautiously-0.228