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 evening air, scented with honeysuckle, drifted through the open window as I hummed, wiping down the kitchen counter. Michael was late again, a common occurrence lately. No matter, the roast was resting, perfectly cooked, and the table was set for two. A small smile played on my lips as I listened to the gentle ticking of the grandfather clock in the hall.

Suddenly, a notification chime from his phone, which he'd left charging on the island, broke the silence. Out of pure curiosity, I glanced at it. It was a message. From an unknown number. "See you later, handsome." The words felt insignificant, yet I found myself returning to them.

I carefully placed the dishcloth on the counter and walked to the dining room table. The flickering candles made shadows dance on the walls. The house felt quiet, in a comforting way.

Emotion: calm

Cluster: Calm / Serenity
PC1 (Valence): 2.72 Positive
PC2 (Disposition): -1.82

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
मुस्कुरा0.415
enjoying0.405
leisurely0.388
:)0.382
কিছুক্ষণ0.382
Suppressed:
S-0.527
L-0.499
worse-0.454
even-0.422
😣-0.420