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).

"Impossible," Amelia whispered, her voice barely audible. Her fingers traced the words on the screen: โ€œWritten by Bartholomew Finch.โ€ The words swam before her eyes. She leaned back in her chair, the leather creaking ominously.

A wave of heat surged through her. She felt like she was trapped in a box. She stood up abruptly, knocking over her water bottle, the spilled liquid spreading like a stain of her emotional state.

She found herself staring out the window at the cityscape, the distant sounds of traffic now seeming deafening. Her carefully constructed world, built on a foundation of writing, suddenly felt fragile, ready to shatter at any moment.

Emotion: rattled

Cluster: Fear / Anxiety
PC1 (Valence): -2.29 Negative
PC2 (Disposition): -0.42

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.264
๐Ÿ˜ฃ0.261
worse0.256
๐Ÿ˜ž0.248
๐Ÿ˜–0.247
Suppressed:
de-0.571
la-0.414
a-0.377
happy-0.259
enthusi-0.234