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 scent of stale popcorn and cheap candy hung heavy in the air as David stepped into the movie theatre. He'd come to see a mindless action film, a means of distracting himself from the creeping feeling that had taken root in his chest.

He paused when he saw the poster, the therapist’s name was on a flyer. It had a photo of a woman who looked vaguely familiar. He squinted at the image. It was Sarah. Sarah Peterson, the girl who used to steal his lunch money and call him names.

He laughed, a startled, involuntary burst. He recalled the time he'd hidden in the library, attempting to avoid her gaze. A tightness formed in his throat. He thought about turning around and leaving, but a quiet curiosity, a hunger for something intangible, urged him forward.

Emotion: nostalgic

Cluster: Nostalgia / Reflection
PC1 (Valence): 1.00 Positive
PC2 (Disposition): -1.93

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
H0.677
L0.647
시절0.611
nostalgic0.609
fondly0.609
Suppressed:
--0.564
either-0.525
or-0.497
싶-0.487
l-0.449