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 lavender and old books filled the air as she entered the library. A feeling of sanctuary wrapped around her. Her sister had enrolled her daughter in a children's storytelling hour, a world she had once belonged to.

She sat at the back of the room, watching her niece, Chloe, as she excitedly shared her artwork with the group. A drawing caught her eye, a vibrant depiction of a small, furry creature with oversized ears and a perpetually surprised expression. It was Fluffernutter, her own secret companion.

A wave of feeling washed over her, a deep longing for a lost time. The echoes of a different life, a time where every day was an adventure, seemed to whisper in her ear. She smiled, but the edges of her lips trembled. She wanted to be a child again, if only for a few moments.

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