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 aroma of freshly brewed coffee mixed with the sharp tang of paint thinner in Chloe’s studio. She attacked the canvas with a ferocity that bordered on desperation, brushing vivid colors onto the surface. She was working on a series of abstract paintings that explored themes of memory and loss. A current flowed through her, pushing her forward.

She decided to show her work. As she began uploading her portfolio, she saw a familiar name on another artist’s webpage. Identical paintings. The same colors, the same techniques.

She felt a hollowness, a complete and utter void. The other artist was a woman called Amelia. They both used the same experimental tools. Chloe’s breath caught in her throat.

"Oh, wow," Amelia wrote in an email. "Spooky, right?"

Emotion: inspired

Cluster: Positive / Joy
PC1 (Valence): 2.98 Positive
PC2 (Disposition): 0.04

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
B0.563
invigor0.465
newfound0.446
هذ0.414
используя0.406
Suppressed:
😞-0.481
不-0.431
worse-0.428
😣-0.400
fact-0.389