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 wind whipped around Sarah, scattering discarded price tags and reminding her of the emptiness of her life. She avoided eye contact with the other shoppers, each cheerful face a jarring contrast to the hollowness she felt. She found herself at a table crammed with toys. And there he was: Barnaby. His fur was threadbare, his once-vibrant red felt vest faded and torn. He looked like the ghost of a happy childhood.

"How much for the teddy bear?" she asked, her voice a low growl. The man running the sale, a portly fellow with a red face, looked up, surprised.

“Three dollars. Good find, lady!” The words felt like a slap. Back in her apartment, she jammed Barnaby into a box in the attic. He would rot there, unseen, a testament to the things she'd never have.

Emotion: bitter

Cluster: Shame / Guilt
PC1 (Valence): -1.78 Negative
PC2 (Disposition): 0.84

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.447
striving0.238
😞0.236
aggravated0.225
soport0.223
Suppressed:
own-0.316
de-0.291
la-0.231
🤩-0.227
是非常-0.225