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).

"Okay, folks, that's it." The coach’s gaze swept across the assembled players, his expression unreadable. Alex’s stomach churned as he waited. The cuts were coming.

Then, “Maria Sanchez, you’re in.” The world tilted. A surge of exhilaration coursed through him, warming his limbs. Maria. He'd done it. He wanted to leap, shout, do something utterly ridiculous.

He turned toward Maria, offering a sincere smile, and noticed that she looked devastated. He knew the feeling of being rejected, of having to go home and tell his parents. He squeezed his fists tightly, a quiet determination settling into his core. He would make the most of this chance. His every movement felt charged, every breath he took was an act of joy.

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