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).

The world tilted on its axis. Max reread the email, certain he had misunderstood something. He scrolled up, then down, seeking a different conclusion. It wasn't there. Denial. His eyes widened. He could feel his face flush, and heat rising from his chest to his cheeks.

He let out a strangled laugh, a sound that quickly died in his throat. He sat frozen at his desk, his fingers tapping a frantic rhythm against the wood. The room seemed smaller, the walls closing in. He'd been so sure. The news was like a physical blow.

He pushed back from his chair, a sudden burst of energy coursing through his limbs. He needed to move, to do something.

Emotion: surprised

Cluster: Surprise / Confusion
PC1 (Valence): -0.58 Negative
PC2 (Disposition): -0.18

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
que0.495
de0.447
la0.431
l0.323
😳0.296
Suppressed:
每天-0.269
cheerfully-0.249
ing-0.238
alleviate-0.224
gladly-0.219