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).

He snatched the envelope from the mailbox, the crisp paper practically vibrating in his fingers. Liam had been expecting this news, but hoping against hope. He found a spot on a bench in the park, sunlight dappling the page as he ripped it open.

He scanned the letter, his eyes widening with each sentence. A smile stretched across his face, and he found himself bouncing on the bench. He couldn't believe it. Pure, unadulterated elation surged through him.

He stood and let out a whoop of joy that startled a flock of pigeons. He didn't care. The world felt suddenly brighter, lighter, as though a heavy cloak had been lifted from his shoulders. He felt like he could run a marathon.

Emotion: delighted

Cluster: Positive / Joy
PC1 (Valence): 3.80 Positive
PC2 (Disposition): -0.26

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
!0.611
🤩0.536
de0.530
ecstatic0.527
🎉0.524
Suppressed:
S-0.687
😞-0.562
ness-0.471
😔-0.385
😥-0.379