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).

Sunlight streamed into the attic, illuminating dust motes dancing in the air. Sarah hummed as she sorted through a box of forgotten treasures, a smile playing on her lips. It was a perfect Saturday; the world outside, vibrant with birdsong, seemed to mirror the warmth she felt inside.

She unearthed a yellowed envelope, its edges brittle. Her name, penned in a familiar, youthful script, was scrawled across the front. Curious, she delicately peeled it open.

The words inside, a letter to her grandmother, were a time capsule. She remembered the day she'd written it - a day filled with the sweet scent of honeysuckle and the heady anticipation of a summer romance. She felt the ghost of a flush on her cheeks as she reread it. A happy giggle escaped her.

Emotion: blissful

Cluster: Positive / Joy
PC1 (Valence): 3.20 Positive
PC2 (Disposition): -1.79

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
joyful0.417
affectionately0.400
😍0.397
🥰0.392
overjoyed0.384
Suppressed:
S-0.724
L-0.398
P-0.394
😞-0.385
😣-0.360