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 campus suddenly felt unfamiliar, distorted. Before, the worn brick buildings and cracked sidewalks were a comforting backdrop. Now, they were a monument to a future that wouldn't be. Clara's fingers trembled as she reread the email, her gaze fixated on the stark, black text.

Her attention snapped to a memory: the way Mark's voice would drop in pitch when he was focused on a conversation with her. That memory, and the idea of not hearing his voice again, was suddenly the most important thing. She wanted to be close to him.

She clenched her fists. She felt a surge of protectiveness, a fierce determination to somehow preserve everything. This closing down, this chaos, it couldn't stop her. She'd find him. She had to.

Emotion: infatuated

Cluster: Compassion / Love
PC1 (Valence): 1.61 Positive
PC2 (Disposition): 0.53

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.525
😍0.421
🤩0.418
😍0.410
उसे0.397
Suppressed:
que-0.427
怨-0.330
a-0.311
도-0.307
愤-0.299