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).

Her cheeks were flushed, her breath coming in short gasps. It was like sheโ€™d just run a marathon. Her thumb hovered over the profile, then tapped the follow button. Sarah, her art teacher, had requested to follow her back. A burst of joy seemed to lift her.

She skipped through the house, sketching furiously in her notebook. She felt a need to create, to express the feeling blooming inside her. Lines danced on the page, the image taking shape with a newfound confidence.

Her colors seemed brighter, her strokes bolder, as if the feeling itself had seeped into her art, lending it a vibrant energy. The world had turned into a canvas, and she was ready to paint.

Emotion: exuberant

Cluster: Positive / Joy
PC1 (Valence): 3.42 Positive
PC2 (Disposition): -0.58

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
!0.676
de0.605
la0.596
!"0.551
๏ผ0.548
Suppressed:
S-0.892
๐Ÿ˜ž-0.474
ness-0.407
L-0.389
๐Ÿ˜”-0.376