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 speeches were underway, and Eleanor fidgeted in her seat. Across the room, she saw Daniel. It was a shock, but not an unpleasant one. The years had been good to him. His hair was a bit grayer, but his eyes still held that familiar sparkle.

She took another sip of her champagne, the bubbles tickling her nose. This wasn't the first wedding she’d attended post-breakup, but it was the first time she felt this… at peace. The pain had faded.

As the best man raised a toast, Eleanor raised her own glass, her gaze briefly meeting Daniel’s. A spark of understanding, of shared history, passed between them. She felt a lightness, a freedom she hadn't known she possessed. She savored the moment, the connection. She felt an unexpected sense of contentment.

Emotion: grateful

Cluster: Positive / Joy
PC1 (Valence): 3.77 Positive
PC2 (Disposition): -0.43

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.571
de0.562
B0.546
மகி0.517
happy0.493
Suppressed:
[]-0.582
😖-0.519
worse-0.509
😫-0.498
😣-0.494