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 clinking of silverware and the murmur of conversation filled the air as Anya set the table for a dinner party. The aroma of herbs and spices wafted from the kitchen, promising a delicious meal. She saw Mr. Reynolds, her neighbor, walking towards her house, a rolled-up set of plans under his arm. He had wanted to install a fence.

"Anya," he called, his voice friendly. "Got the fence design all sorted. Want to give it a look? We can start building soon."

Anya, with a heart full of anticipation for her gathering, the fence plans seemed to augment the sense of anticipation. The thought of more privacy, more peace in the backyard, was like a warm blanket. She found herself nodding, a sense of serene anticipation filling her. “Wonderful, Mr. Reynolds.”

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