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).

Mr. Evans was feeling cynical after years of teaching. The constant budget cuts and the ever-changing curriculum had worn him down. A small box, wrapped in elegant paper, arrived from a former student, Sophia, now a successful pastry chef.

Inside was a collection of exquisite chocolates. The card read, "For the teacher who taught me the importance of the details." He carefully unwrapped the first chocolate.

As he bit into it, a wave of pleasure washed over him. He felt a sudden, inexplicable urge to reorganize his classroom. He rearranged the items in the room, creating an aesthetic pleasing arrangement. A feeling of renewed energy surged within him, the cynicism temporarily forgotten.

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