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 award show was about to start, and Sarah felt like she could fly. Her design for the new skyscraper had been a long shot, but she knew she'd nailed it. She was sure she had a chance. She was filled with an excited energy, her feet barely touching the ground.

She entered the ballroom, her eyes scanning the room. Then, she stopped. Across the hall, a competitor was on stage, receiving the award. His design was identical to hers.

Her jaw dropped. She felt a wave of dizziness wash over her. A dull ache settled in her chest. All her effort, all her energy... for nothing. She felt the tears well up in her eyes.

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