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 airport buzzed with disgruntled passengers. My grandmother’s birthday party. I'd missed it again. This time it was a mechanical fault. I slumped into a chair, feeling the familiar sting of disappointment. My family deserved better. I should have planned for this.

Then, a small, elderly woman sat next to me. She was knitting, her hands moving with incredible skill. She offered me a piece of hard candy. I took it, the sweetness strangely comforting. She started telling me a story about her own delayed flights, how they led to unexpected friendships and opportunities. As I listened, the edge of my disappointment seemed to soften, replaced by a sense of calm. The delay had offered a chance to connect with someone. It was like a little gift.

Emotion: thankful

Cluster: Positive / Joy
PC1 (Valence): 3.50 Positive
PC2 (Disposition): -0.83

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
happy0.493
மகி0.487
grateful0.461
mutlu0.452
feliz0.431
Suppressed:
[]-0.482
😖-0.478
😣-0.467
😫-0.461
worse-0.450