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 small canvas, her creation, hung in the window of a shop that sold vintage clothing. It was a painting she had abandoned years ago, a piece she had considered a failure. Now, here it was, perfectly positioned to catch the eye of passers-by. She felt a knot of mixed emotions tighten in her stomach.

She stepped inside. The shop owner, a woman with bright red lipstick, greeted her with a warm smile. "Isn't it beautiful?" she gushed, gesturing towards the painting. "We found it tucked away in an old trunk. It's truly a treasure."

She nodded, unable to speak. The light fell on the canvas in a way that made it glow, bringing out the depth and vibrancy she had never seen before. She felt a strange surge of… pride? She wanted to know who had found it. To thank them. Her gaze flickered to the shop owner, who was now ringing up a customer.

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