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).

"Oh, my God," she whispered, her voice barely audible. Her stomach was twisting, and she felt a prickling sensation all over her skin. The website, displaying her meticulously penned diary entries, seemed to pulse with an invasive energy. She ran her fingers through her hair, a frantic gesture of disbelief. It was an invasion, a violation.

The comments, though a blur, were a chorus of voices dissecting her past. She felt pinned, as if she were a butterfly impaled on a board. She felt as though she might simply disappear, fade away. She felt a strange buzzing behind her eyes.

The room seemed to shrink. She took a deep breath, fighting the urge to run. Her younger self, so vulnerable and raw, was being judged by strangers. She felt a cold dread settle in her gut.

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