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).

The air around John seemed to crackle. He had been casually browsing, killing time, when *it* popped up. The house. His childhood home. The photos, though professionally taken, couldn't hide the familiar imperfections. The slightly crooked mailbox, the cracked pavement, the overgrown rose bushesโ€ฆ

He took a deep breath, his chest filling with an emotion he hadn't felt in a very long time. He could feel his hands tremble slightly as he reached for the phone.

โ€œHey, Mom,โ€ he said, his voice unusually high-pitched. He felt like he wanted to jump up and down. He felt like he wanted to scream. "You are not going to BELIEVE this..."

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
๐Ÿคฉ0.499
๐ŸŽ‰0.457
Suppressed:
S-0.601
๐Ÿ˜ž-0.536
ness-0.462
๐Ÿ˜ฅ-0.351
๐Ÿ˜”-0.341