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).

"Again?" The woman's voice, muffled through the door, was almost a whisper, but still clearly audible. Mark practically buzzed with energy. He bounced on his heels, a wide grin plastered across his face. He’d known it wouldn’t be long.

He took a deep, steadying breath. "Just a few more minutes," he promised in a deliberately jovial tone. "Almost there!"

Inside, the instruments of his chaos stood poised. His latest experiment, a symphony of industrial clangs and shrieks designed to disorient and overwhelm, was reaching its crescendo. He rubbed his hands together, the tips tingling. He couldn't wait.

Emotion: thrilled

Cluster: Positive / Joy
PC1 (Valence): 3.31 Positive
PC2 (Disposition): 0.07

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.613
!0.544
B0.508
🤩0.499
🎉0.457
Suppressed:
S-0.601
😞-0.536
ness-0.462
😥-0.351
😔-0.341