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 last box of Christmas decorations was tucked away in the attic. It was a chore she didn't mind, a feeling of contentment bubbling within her. Their life together was vibrant and full of love. She felt lucky. He felt loved. The world was good.

She paused, spotting a small, leather-bound notebook tucked amongst the wrapping paper. It wasn’t hers. It belonged to her husband, Kevin. Curious, she opened it. The first page contained a handwritten note. It detailed a savings account.

The balance was significant, a surprise. He had never discussed a secret savings account. Her heart fluttered. She knew Kevin’s passion for vintage cars. She couldn't stop herself from smiling.

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