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 paced his small apartment, the worn rug under his feet barely registering his movement. The email glowed on his laptop screen: "Your results are in!" A giddy laugh escaped his lips. He'd waited weeks, biting his nails, for this moment. His adopted parents had always been wonderful, but a part of him had always felt incomplete, like a missing chapter. Now, he felt, was the time to begin.

He picked up the phone, dialing his best friend. "Dude," he began, his voice barely a whisper, then grew to a shout, "Guess what? I'm not a Johnson!" He had to tell someone, *anyone*. He felt a light, almost weightless feeling, as if he could float right up to the ceiling. The need to start searching for his birth parents was practically bursting from him.

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