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 scent of lavender wafted through the pottery studio, mingling with the earthy aroma of wet clay. Clara’s hands moved with practiced ease, shaping a beautiful vase on the spinning wheel. The rhythmic hum of the machine, the feel of the clay, it all centered her. A deep satisfaction bloomed within her.

A curt email arrived, a bureaucratic missive from the clinic. Subject: "Important Record Clarification." It described symptoms she'd never experienced and listed medications she’d never taken. The whole thing was bizarre. It felt like a novel.

Clara paused, a slight frown creasing her brow, but then she chuckled. "Well, that's certainly not my life," she murmured, a gentle smile returning. She felt not a shred of fear. A sense of detached curiosity washed over her, and she went back to her pottery.

Emotion: fulfilled

Cluster: Positive / Joy
PC1 (Valence): 3.22 Positive
PC2 (Disposition): -1.74

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.819
மகி0.415
happy0.397
grateful0.389
de0.385
Suppressed:
S-0.599
worse-0.465
C-0.460
either-0.436
😣-0.426