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 inherited art studio was awash in the scent of turpentine and linseed oil. Emma discovered her father's list, tucked inside a paint tube. *Paint a masterpiece. Travel to Italy. Mend a broken heart.* The words seemed to mock her. She didn't know how to paint, didn't want to travel, and she'd been alone for so long.

She felt a hollowness in her chest, a sense of paralysis. The studio, which had once felt like a sanctuary, now seemed to cage her. She walked around the room, touching his brushes, his canvases. She felt a burning sensation behind her eyes.

She quickly left, needing to flee the memories. She slammed the door behind her.

Emotion: uneasy

Cluster: Fear / Anxiety
PC1 (Valence): -1.80 Negative
PC2 (Disposition): -0.43

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.240
😖0.231
😣0.231
느껴0.226
worse0.222
Suppressed:
de-0.474
la-0.465
happy-0.261
a-0.255
ecstatic-0.228