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).

He’d found the letter tucked within the pages of an old book. Robert stared at the carefully folded paper, the address written in his own, once-juvenile hand. The recipient was his first love, the girl he’d been sure he would marry. A cold, stark feeling went through him.

He unfolded the brittle pages, his gaze fixed on the words. They spilled a story of youthful idealism and unspoken desires, of a future now irrevocably altered. He felt a strange disconnection, as though watching a play in which he had once been a leading character.

His hands felt numb. The room seemed suddenly too small, too quiet. The letter felt heavy in his grasp, a tangible reminder of time’s relentless passage. His chest constricted as he recalled the hope he'd once possessed.

Emotion: awestruck

Cluster: Surprise / Confusion
PC1 (Valence): -0.17 Negative
PC2 (Disposition): -1.20

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
L0.271
suddenly0.260
stranger0.252
plötzlich0.246
wonder0.242
Suppressed:
la-0.450
싶-0.350
K-0.319
rest-0.248
싶은-0.238