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).

"This is ridiculous," Michael muttered under his breath, swatting at a cobweb. The attic was a disaster, a repository for the previous residents' unwanted possessions. He had delayed the inevitable for far too long. Now, he had to face the mountain of accumulated things.

He uncovered a box containing a collection of amateur paintings. Landscapes, portraits, still lifes—each one a testament to someone’s artistic ambition. He held up a canvas depicting a sunset, its colors faded and muted. He didn't even know where to begin.

His gaze flicked around the room, taking in the disorder. He should have dealt with this weeks ago. A sense of inadequacy settled over him. He felt as though he was failing at the simplest of tasks. He ran a hand through his hair and wished he was back downstairs.

Emotion: ashamed

Cluster: Shame / Guilt
PC1 (Valence): -1.97 Negative
PC2 (Disposition): 1.06

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.376
ashamed0.344
ness0.327
逃0.297
😞0.295
Suppressed:
de-0.539
que-0.300
😍-0.288
(!)-0.269
🤩-0.242