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 spilled across Amelia’s kitchen, illuminating the dust motes dancing in the air. She hummed off-key as she flipped a pancake, a grin plastered across her face. The book, a musty copy of "Pride and Prejudice," lay open on the counter, propping up the syrup bottle. This morning, she’d felt an urge to be extravagant, so she had prepared a full breakfast spread, complete with fresh fruit and a pot of Earl Grey tea.

Suddenly, a crinkling sound. A small, yellowed envelope slipped free from the pages. Amelia’s pulse quickened. She carefully extracted the letter, a flush warming her cheeks. The script was delicate, looping, and the words practically sang on the page.

She recognized the signature: 'Yours always, Elizabeth.' A thrill zipped through her as she realized she was holding someone’s hidden secret, a moment of another life, a connection to a past she now had a part in.

Emotion: vibrant

Cluster: Positive / Joy
PC1 (Valence): 2.83 Positive
PC2 (Disposition): -1.36

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.518
!0.459
!".0.436
de0.434
!).0.409
Suppressed:
S-0.890
L-0.478
P-0.407
H-0.365
😞-0.306