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 news felt like a fizzy drink bubbling inside Elias. He could barely contain his excitement. He'd done it! He'd been accepted! He'd spent the last year studying, working, and sacrificing, and it had all been worth it. The college acceptance letter lay on the table.

He started composing a celebratory text message to his friends. The college town was just a few hours away, and he could barely wait to experience it. He decided to walk to the park to calm his nerves.

As he was walking, he decided to check his email. He'd almost forgotten the scholarship application. Then he saw the notification. With a sinking feeling, he clicked on it. The words swam before his eyes. He stopped dead in his tracks.

Emotion: exuberant

Cluster: Positive / Joy
PC1 (Valence): 3.42 Positive
PC2 (Disposition): -0.58

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
!0.676
de0.605
la0.596
!"0.551
!0.548
Suppressed:
S-0.892
😞-0.474
ness-0.407
L-0.389
😔-0.376