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).

His presence made the room spin. Oliver saw Ava across the dance floor. The happy music seemed to grate on his ears, each joyous note a hammer blow to his composure. He felt pinned down, exposed.

He took a deep, unsteady breath. He'd thought he was past this. The sting of the breakup was supposed to be a distant memory. He felt a hot flush creep up his neck, spreading across his face.

He fiddled with his tie, trying to appear unconcerned. He mumbled an excuse to his companion and made a beeline for the nearest exit, his heart hammering against his ribs. He just needed to get away.

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