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).

It was the lilies of the valley that did it. Delicate, elegant, the same flowers he'd given her years ago, a lifetime ago. He couldn't help the gasp that escaped him when he saw them on the doorstep. His fingers went numb.

He stumbled backwards, knocking over a potted plant. The soil spilled across the porch. The sudden mess was a poor reflection of the chaos in his thoughts. The lack of explanation, the subtle hint, left him paralyzed.

His breath hitched as he stared at the blossoms. He felt a cold sweat break out on his forehead. It couldn't be. Could it? He backed away slowly, as if the flowers themselves were a threat.

Emotion: shocked

Cluster: Surprise / Confusion
PC1 (Valence): -2.19 Negative
PC2 (Disposition): -0.65

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
будто0.298
😰0.268
sickening0.266
не0.266
que0.257
Suppressed:
enthusi-0.283
开心-0.269
ولة-0.262
ارك-0.257
উৎসাহিত-0.255